2
Robotic Arm & Dexterous Hand Senior Design
[pic]
Critical Design Report
Project 05911
David Parrett
Wen Jia Wang
Justin Tubiolo
Jeremy Amidon
Ken Peters
Executive Summary
The purpose of this report is to describe the design process of an interactive robotic arm and hand to be placed in a museum environment for the enjoyment of children. It was the intention of the Rochester Museum and Science Center (RMSC) for the team to propose various exhibit ideas that utilize robotics and then to develop one of these ideas into a working exhibit for actual display and use in the museum. With a large emphasis on the word "fun," the team set about to find a creative and inspiring idea that would not only appeal to children but also serve to enhance their creativity and learning. The end result was the concept of a robotic arm and dexterous hand which the children can control through use of a wearable glove that tracks the movements of their hand, arm, and fingers. The robotic arm and dexterous hand are used inside of an enclosed area to move, lift, and otherwise interact with various objects inside of the display area. The main goal for the team has been to allow children to successfully transfer their movements to the movements of the robotic arm and hand while keeping it fun and exciting for the children.
By using the Engineering Design PlannerTM methodology, the team was able to design the Robotic Arm and Dexterous Hand through five different facets.
The first facet of the design is called Recognizing the Need and Defining the Problem. The team has spent a large part of this past design period conferring with the RMSC to establish the needs for this project. Issues such as safety, repair, interaction, and learning have all been discussed along with the usual issues of normal operation. In order to choose the project, a list of project proposals was created and submitted to the sponsor. Once the project was decided and key issues and objectives had been defined, it was important that the team propose many different concepts for the museum to choose from and then to further develop the final project idea into a list of necessary components and activities. This Concept Development is included as the second facet of this report. The third facet, Feasibility Assessment, is concerned with analyzing the different design alternatives to meet a specific need in the design. Choices such as whether to design and manufacture a specific component or to buy one pre-manufactured are considered while weighing the costs and benefits of each decision. The resulting design choices then allowed the team to generate the fourth facet, Specifications for how exactly the robotics will mechanically and electrically work. Using these decisions and specifications, the team then took the final step in the design process, Analyses and Synthesis, to finalize and fill in the design details including technical drawings and algorithms.
.
Executive Summary 2
1.0 RECOGNIZE AND QUANTIFY NEEDS 6
1.1 Mission Statement 6
1.2 Project Description 6
1.3 Scope Limitations 8
1.4 Stakeholders 8
1.5 Key Business Goals 9
1.6 Top Level Critical Financial Parameters 10
1.7 Financial Analysis 10
1.8 Market 11
1.9 Order Qualifiers 11
1.10 Order Winners 11
2.0 PROJECT PROPOSALS 12
2.1 Robotic Arm and Dexterous Hand 12
2.2 Robotic Squirt Gun and Targets 13
2.3 Remote Racing and Track 15
2.4 Robotic Arm Wrestling 16
2.5 Remote Soccer Players 18
2.6 Robotic Basketball Shooter 19
2.7 Interactive Maneuverable 3-D Maze 20
2.8 Conclusion 22
3.0 Concept Development 23
3.1 Controls 23
3.1.1 Sensor Laced Glove & Fighter Style Joystick 23
3.1.2 Sensor Glove with Motion Track 24
3.2 Arm Types 25
3.2.1 Wrist, Elbow, & Shoulder Joints 25
3.2.2 Wrist, Elbow & Rotating Cylinder Joints 26
3.3 Source of Mechanical Power 26
3.3.1 Pneumatics 26
3.3.2 Electric Motors 27
3.3.3 Hybrid System 28
3.4 Electronic Data and Control System 29
3.4.1 PC with an output board 29
3.4.2 Microcontroller Development Board Alone 29
3.4.3 PC with a Microcontroller Development Board 30
4.0 Feasibility 31
4.1 Project Feasibility 31
4.2 Controller Feasibility 32
4.3 Arm Feasibility 33
4.4 Electronic Control Feasibility 34
5.0 OBJECTIVES & SPECIFICATIONS 35
5.2 Design Objectives 35
5.2 Performance Specifications 36
5.3 Design Practices 36
5.4 Safety Issues 37
6.0 DESIGN ANALYSIS & SYNTHESIS 38
6.1 Display Analysis & Synthesis 38
6.2 Hand Design Analysis and Synthesis 39
6.3 Forearm Design Analysis and Synthesis 41
6.4 Upper Arm Design Analysis and Synthesis 42
6.5 Equations Used for Mechanical Design Analysis and Synthesis 43
6.6 Electrical Control 44
6.6.1 Electrical Input and Output 44
6.6.2 Electrical Control Algorithm 44
6.6.3 PC Program in C++ 45
6.6.4 Microcontroller Program for 8051 MCU 45
7.0 DELIVERABLES 46
7.1 The Robotic Arm and Hand Display 46
7.2 The Robotic Arm and Hand User’s Guide 46
7.3 The Complete Parts List with Vendors 46
8.0 PROJECT TIMELINE 47
9.0 PROJECT BUDGET 48
1.0 RECOGNIZE AND QUANTIFY NEEDS
1.1 Mission Statement
The purpose of this senior design team is to create a robotics display for the RMSC which is fun, inviting, accessible, and easy to understand and use. The new display will be a robotic arm and dexterous hand which will copy the movements of the user's hand in a glove controller. The robotics display will be suitable for placement and use on the museum floor as a working exhibit to be used especially by children ages 8-14.
2 Project Description
Robotics is a field of technology in which there is a great demand and natural curiosity to explore what intelligent machines can do to help people. A robot can be either very intelligent and responsive to its natural environment, or it can perform a set task without any intelligent response. In this project it is important for the safe and normal operation of the robot, that it respond to its environment by detecting when it has reached its fullest extent of motion or has collided with an immovable object to cease movement in that direction. It is also important for the robot to maintain a smooth, steady motion which corresponds to how the user desires the robot to move.
Besides simply moving around and mimicking the motion of the human hand and arm, the robot will also be capable of performing three or four tasks which will present a small challenge and add interest to the display. It was mentioned by the RMSC that the arm should perform activities that could not easily be done by the children with their own hands. One possible example of this is to allow the robot to pick up a basketball with one hand and place it in a basket. Another possibility is to have the robot pick up a recognizably heavy object such as a construction brick and be able to stack several of them together. Other possible activities could be placing odd-shaped objects through corresponding holes, and touching the robotic hand to a target where the hand will conduct electricity to turn on a light.
The basic shape and layout of the exhibit is demonstrated in Figure 1-1. The arm is shown supported from above the display table where the various objects for interaction will be located. The arm features rotational movement at the mounting point to allow it to swivel from side to side. Just below this rotational motor is located a cylinder which allows the arm to extend and retract give the arm variable reach. Below the extension cylinder is the elbow joint for the arm allowing it to bend in various life-like postures. On the forearm section below the elbow joint will be placed air muscles which will allow the wrist and individual fingers of the hand to bend.
The control area is located in front of the display case and houses the glove controller. In order to save wear and tear on the controller and allow the glove to operate properly with the sensor, it is necessary to surround the glove on all but one side to limit the user's freedom of movement. The limitation allows the user enough freedom to move the glove in all directions for between one to two feet from center and keeps the glove on a restraining leash to make sure it remains with the exhibit.
1.3 Scope Limitations
One major idea which the team is striving for is to avoid reinventing and remanufacturing pre-existing design components. It is necessary for the team to purchase working components in nearly all cases to keep the cost low and provide the working prototype by the established deadline. Though it would be possible to completely design and manufacture the project from scratch, especially the glove controller, this would require much more design, manufacture, and test time than is available for this project.
The team is limited, to a degree, by the $5000 budget but the team does not believe that the project will need to cost more than this. Responsible spending and good component choices are important to staying within this budget constraint. The RMSC has been very supportive in promising facilities such as a compressed air line for pneumatics components and has showed a desire to avoid placing creative limits on the project in general wherever possible.
The major constraint of the project is the schedule which is set to a strict timeline with two main deadlines. At the end of the fall quarter, 20041, the team must deliver the detailed design package, quotes for vendors, and the proposed budget. By the end of the winter quarter, 20042, the team will provide the working prototype, the final report, and test results of both the operational specifications and human interactive requirements.
1.4 Stakeholders
The main stakeholders in this project are the project sponsor, Rochester Museum and Science Center, and the senior design team members. The Museum has a large stake because of the possibly huge attraction and technologically-advanced image that this project could bring to the RMSC should it be integrated successfully as an exhibit there. The senior design team members also have stake in this project as a demonstration of their ability to solve engineering problems and gain invaluable experience in the expanding field of Robotics. The team's faculty advisors and the college of engineering also has stake in the project because of the potential for future sponsorship of senior design projects by the RMSC. Outside vendors are also stakeholders which the team shall be purchasing many of the commonly manufactured parts for integration in the design. More stakeholders include the members of future RIT robotics senior design teams, other schools doing research on miniature turbines that could benefit from our results, any other designers and users of publicly exhibited robotics, and the future employers of the team members.
1.5 Key Business Goals
A successful project will be defined by the evidence of a working robotic arm and dexterous hand prototype achieving full dexterity, range of motion, and control by the user to manipulate the objects inside the display. If the design team is capable of completing such a task, then much will have been accomplished. Not only will the core objectives of the project be achieved, the students on the team will have also gained a valuable experience in working within a multidisciplinary team. The results of this project will serve as a tangible incentive for further development by museums in robotics displays. Success of this project will bring future opportunities for attracting a wide group of interested young people into the museum, and possibly eventually into the study of engineering and the field of robotics.
1.6 Top Level Critical Financial Parameters
The financial needs of this project are driven primarily by the purchase of the individual components that will be assembled for the mechanical robotic function. The major electrical system contributors to the overall cost include a personal computer and microcontroller development board. The major mechanical cost contributors include the pneumatic system, including air muscles, air lines, and valves, along with the mechanical joints, motors, and extension cylinder. There may also be small fees due to persons external to the team, who are responsible to manufacture various mechanical components but these will be relatively small in comparison to off-the-shelf mechanical purchases.
1.7 Financial Analysis
The team is working with a $5000 dollar budget cap in mind. The RMSC will work with the team to determine which possible budget items can be donated and which will need to be purchased. The plan for spending is to purchase the major components of the design that are absolutely necessary by the Preliminary Design Review and at an early stage of assembly in the following quarter to ensure that they will be properly integrated and to allow for a fallback period if necessary. The $5000 dollar limit is deemed by the team to be sufficient for this project; however there are several design options which, if chosen, would add to the cost significantly. The electrical components will require an estimated five hundred to seven hundred dollars and the majority of funds will go to the mechanical side of the project including, but not limited to:
• Pneumatic component cost: air muscles, valves, and lines
• Mechanical control components: extension cylinder and electric motors
• Material and machining cost: hand and finger joints
1.8 Market
The Robotic Arm and Dexterous Hand Display is intended to be used exclusively by the RMSC. Though the design and concepts used throughout this project will be available to the RIT community and could prove useful to similar projects in the future, it is not intended that this project be reproduced for public marketing and distribution.
1.9 Order Qualifiers
The purpose of this research design is to provide a fun and interactive display for use in a museum exhibit by children. The robot needs to satisfy this requirement by allowing the user to control it simply and easily which will translate to an efficient and pleasing operation of the robotic mechanism. The operation must involve the ability to move fingers, hand, and arm throughout the display unhindered, and it must allow the user to interact with the objects inside the display in an intuitive manner.
1.10 Order Winners
If time and money permit the team will work to complete the following goals:
• Design and build to ensure maximum exhibit life span and durability.
• Provide for intuitive navigation and operation by users.
• Allow the robotics theme to be an exceptionally inviting "attractor factor."
• Provide for accessibility to many users of different age and physical ability.
• Provide maintenance information such as drawings and replacement parts.
• Make the user feel as though the robotics exhibit is an extension of them self.
• Test the display in its target museum environment to ensure its success.
PROJECT PROPOSALS
The RMSC requested that the team brainstorm and propose various robotics display ideas to be considered for this project. The ideas were in many different directions but always centered on being interactive and inviting to children. The following are the seven project ideas which were proposed.
2.1 Robotic Arm and Dexterous Hand
2.1.1 Overview
This design concept incorporates a robotic arm with a fully functional robotic hand. The arm and hand could be used to complete various tasks like stacking blocks or lifting and moving heavy objects. The robot would be contained in a display case that would be easily viewed by the operator as well as spectators.
2.1.2 The "Attractor Factor"
The sight of the robotic arm and hand should be a significant draw. Also, the controls will draw people to try the glove and joysticks.
2.1.3 Method of Interaction
The arm and hand would be controlled by sensor-laced glove and one to two joysticks. Patrons would put the glove on and the robotic hand would mimic the movements of their hand. The joysticks would be used to control the motion of the robotic arm. We would explore the possibilities of incorporating different size gloves like a small, medium, and large to increase the range of ages allowed to operate the robot.
2.1.4 Scientific Learning
We would like to look into the possibility of smaller display cases around the outside of the robot case that would contain various parts and equipment used in the robot with explanations as to what they are for. These display cases would allow people observing the robot to see what types of technologies went into making the robot and hopefully learn a little about robotics.
2.1.5 Intuitive Operation
This display would not require elaborate instructions before using because you would learn to operate it by playing with it.
2.1.6 Materials
The mechanics of the robot would most likely be comprised of steel and aluminum components to ensure durability. There would be a multitude of electronic sensors, circuitry and motors to control the robot. We are looking into the different methods of creating the motion of the robot including electric, hydraulic and pneumatic.
2.2 Robotic Squirt Gun and Targets
2.2.1 Overview
The squirting hand is an idea which incorporates a robotic hand with a squirt gun. The squirting part would be built into one of the fingers. In the display would be several targets and other things that kids could shoot water at. One idea was to put a turbine which they can shoot at and light a light bulb. They could then experiment with shooting the turbine in different places and seeing which spins it faster, lighting the bulb brighter. This idea is also a possibility to combine with the sensor/glove idea. Basically, I envision a display case with the hand on one side. The controls will be behind the hand and in front will be a “shooting range” with whatever targets we would choose.
2.2.2 The "Attractor Factor"
The attractor factor of this project lies in the fact that kids can shoot things. All kids love squirt guns and would find this enjoyable. Adding things like light bulb to light up would help draw kids to this display.
2.2.3 Method of Interaction
Through controlling the hand, and squirting device, either through joystick or glove allows the kids to interact with the display.
2.2.4 Scientific Learning
This project includes scientific learning through the turbine, teaching them about circular motion. Other possibilities are linking an electricity lesson to the light bulb or a lesson on fluids to the squirting mechanism.
2.2.5 Intuitive Operation
This particular display would require little to no instruction.
2.2.6 Materials
Materials would include the robotic hand and whatever controls are associated with it. We would also have to build the “shooting range” and whatever targets we desire. The display would have to funnel the water back to some sort of pump to be reused.
2.3 Remote Racing and Track
2.3.1 Overview
The race track idea is pretty much a high-tech racing game. Picture a regular electric racing game but add some intelligence. Some cars could be computer controlled allowing from 1-8 players. Players would be penalized for bumping another car, running into the wall, etc. Built in could be noises, fires, and smoke, associated with certain occurrences. Here we would have a large oval racetrack. With eight control stations set up around the circumference, each with its own wheel and pedals. Brightly painted cars would sit on the track along with other effects like spectators in bleachers, making it look like a real racetrack.
2.3.2 The "Attractor Factor"
The attractor factor is that a big race track would be there, with sounds, smoke, etc. Kids would see the cool racecars and want to try it out. Controls could be separate stations with gas and brake pedals and a steering wheel.
2.3.3 Method of Interaction
Through controlling the cars, and being involved with a race, kids would interact with the display.
2.3.4 Scientific Learning
This project includes a little scientific learning but lots more fun for kids.
2.3.5 Intuitive Operation
This display would require a little instruction in using the remote car controls.
2.3.6 Materials
Materials would include the track, the cars and the control devices.
2.4 Robotic Arm Wrestling
2.4.1 Overview
The Robotic Arm Wrestling Project will focus on the physical capabilities of robotics. There will be two robotic arms placed in a display. These arms will be positioned such that they look like two people arm wrestling. The arms will be hinged at the bottom, or the elbow, with joints at the wrist and fingers. Each arm will be controlled by a different operator at opposite ends of the display. In order to make your arm push down on the opposing arm, the operator must alternate pushing two buttons as fast as possible. The faster the buttons are pushed, the harder the operators arm pushes down on the opposing arm.
2.4.2 The "Attractor Factor"
Robotic Arm Wrestling would be an extremely attractive display for several reasons. First, the mere aspect of a robotics display would attract a child. Robotics is very unique and very enticing for children. Second, arm wrestling is something that many children do with their friends all the time. After seeing a display that involves arm wrestling a child would be more likely to come and try this display. Another reason Robotic Arm Wrestling has the “attractor factor” is that it involves two operators. A child won’t just be playing this game by him/herself. Children can actually compete with one another, which is always attractive to children.
2.4.3 Method of Interaction
Children will get to physically control the robotic arms in order to generate movement by pushing down repeatedly on alternating buttons. As they press these buttons they can see the movement of the arms in a real time atmosphere.
2.4.4 Scientific Learning
We would like to look into the possibility of smaller display cases around the outside of the robot case that would contain various parts and equipment used in the robot with explanations as to what they are for. These display cases would allow people observing the robot to see what types of technologies went into making the robot and hopefully learn a little about robotics.
2.4.5 Intuitive Operation
The Robotic Arm Wrestling display will require some direction. As with almost any hands on display, there will need to be a description of what the operator needs to do in order to accomplish the objective. The ideas of this project are very simple, which will require minimal direction and supervision. It does not take much to explain that the operator will need to alternate pressing two buttons in rapid succession in order to control the arm.
2.4.6 Materials
The materials would include metal for arms and fingers, joint systems, electrical system, and display case and buttons.
2.5 Remote Soccer Players
2.5.1 Overview
In this project we will design at least two small robots. Each autonomous robot will have a remote control, four wheels to drive, and two “tools” to play soccer with. The first tool is a “leg” which can pass the ball to other team members. The second tool is an “eye.” Each robot will have an eye and a small light bubble. When a robot's sensor detects a light, it will stop a while to give a chance to the other team to get the ball. Each robot would be as small as possible for maneuvering and have as long a battery life as possible. There will also be a miniature soccer field and goals at each end to simulate the soccer environment.
2.5.2 The "Attractor Factor"
Kids always enjoy remote control cars and this will give them an opportunity to control robotic soccer-kicking miniature vehicles.
2.5.3 Method of Interaction
The user would press a particular button to kick the ball and would control the vehicles with a remote control. It might be necessary to have separate stations for different vehicles similar to the remote racing.
2.5.4 Scientific Learning
The soccer robots will teach children about intelligent robots today that sometimes will act without human control. This is to let them know that modern robots can “feel” and that they are not just machines. In the future they will respond more and more like humans.
2.5.5 Intuitive Operation
This display would require a little instruction in using the remote car controls.
2.5.6 Materials
Each robotic vehicle would require a light bubble, light sensors, motors, program chip, mechanical foot, and a vehicle frame and wheels. The display would require materials for a goal, playing field, and protective boundary.
2.6 Robotic Basketball Shooter
2.6.1 Overview
The robotic basketball shooter will mimic the action of the human arm while throwing a basketball into a basket. The control would be three simple levers which turn the shoulder, elbow, and wrist of the robotic arm. The challenge is to properly time how the wrist and elbow move so that the robot will correctly throw the ball. It would also be possible to change skill levels. The display will involve an enclosed area in which to throw the basketball and allow it to be retrieved automatically and replaced in the robotic hand. When the user moves a lever, the change is immediately translated to a movement in the robotic arm. One possible alternative is to use an arm like this to throw a smaller ball through a target that will light up or react when the toss is successful.
2.6.2 The "Attractor Factor"
The simple controls and robotic arm will definitely draw spectators. The basketball theme would also help to draw people when they see the basket.
2.6.3 Method of Interaction
The control would be three simple levers as mentioned above. The shoulder lever would be used to line up the shot and the other two would be used to throw the ball.
2.6.4 Scientific Learning
There are a broad range of opportunities for children to learn how their arm moves when they throw a ball and how robots can try to reproduce that motion.
2.6.5 Intuitive Operation
This display might require a little explanation but even without any, a child will quickly learn how it works by playing with the levers and watching the arm move.
2.6.6 Materials
The arm itself would include metal and other sturdy construction components to allow for the stress placed by throwing a ball. Three motors for the joints and control circuitry for each from the input levers would be necessary along with additional mechanics to automatically retrieve and place the ball in the robotic hand. The display would involve a sturdy, transparent barrier for safety, a basketball backboard and rim, and a control table where the levers would be mounted.
2.7 Interactive Maneuverable 3-D Maze
2.7.1 Overview
The Maneuverable Maze is a concept similar to common, smaller versions of a maze which tilts in all directions to allow a ball to roll through from the starting place to the end (sometimes called a "labyrinth"). The maze would be mounted on a robotics base to allow it to tilt in all directions. The actual maze would be quite large and the motion control system would be a small version of the maze mounted on a similar support structure. When someone moves the miniature maze controller, the large maze would copy that motion and the ball would move around inside the maze. It would be possible to include different obstacles and interaction within the maze to increase or decrease the challenge of finishing the puzzle. The control device would be located on an elevated platform to allow easy viewing of the large maze below which would be surrounded by a protective display case. There could also be the addition of moveable "doors" or obstructions inside the maze that would be controlled by the user to increase their interaction with the maze.
2.7.2 The "Attractor Factor"
The large maze would certainly attract attention and the opportunity to turn and tilt such a large maze with a smaller version would help to hold their attention. This project also has the benefit of allowing modification to the average patron "dwell time" by changing the difficulty.
2.7.3 Method of Interaction
The controller would be a moveable smaller version of the larger maze on a robotic/hydraulic pedestal. The large maze would immediately copy how the user changes the position of the controller.
2.7.4 Scientific Learning
The maze is more of a fun puzzle-solving experience which utilizes robotics than a learning experience. A basic understanding of gravity is certainly necessary to predict how the ball will roll through the maze, but the emphasis here is mostly on fun and challenging interaction with the robotics.
2.7.5 Intuitive Operation
This display would require very little explanation to tell users to move the small maze and watch the large one move the same way.
2.7.6 Materials
An elevated platform, display case, control device/pedestal, robotic support equipment for both the control device and large maze, control electronics, the large maze made of metal/plastic/wood, and the ball.
2.8 Conclusion
All of the project proposals were presented to the Rochester Museum and Science Center to obtain their feedback on what they saw as the best fit for their facility. After multiple meetings with Museum staff the projects were narrowed down to three, the basketball shooter, the interactive maze and the robotic arm and dexterous hand. In the end the robotic arm and dexterous hand was selected because it was the best example of robotics and the most interactive for museum patrons.
Concept Development
The concept development for the project considered the areas of control mechanism, the orientation and operation of the robotic arm, the mechanical power source to move the arm, and the electrical system controlling the movement.
3.1 Controls
3.1.1 Sensor Laced Glove & Fighter Style Joystick
One concept for controlling the arm and hand is to utilize a glove laced with flex sensors on one hand and operating a fighter style joystick with the other hand. The glove would have seven flex sensors embedded in it to sense the position of the operator’s fingers, thumb and wrist. The sensors work by changing their resistance the more they are bent. This change in resistance could be calibrated to represent the position of the finger.
Courtesy of
The data obtained from the glove would be used to control the fingers and wrist so the robot would in a sense mimic the operator’s movement. The joystick would be used to control the other parameters of the arm. Most likely moving the joystick left and right would control the swivel and moving it forward and backward would control the elbow joint. The cylinder would be controlled by the trigger and thumb buttons on the joystick.
Courtesy of Radio Shack
2 Sensor Glove with Motion Track
This concept would utilize a similar glove as concept 3.2.1 but the glove’s motion would also be used to control the rest of the arm. Basically the robotic arm would mimic all the operator’s movements. Initially this was our ideal way of controlling the robot but we didn’t think it was feasible for our budget and resources. That was of course until we found a glove on the market that we could purchase to accomplish this. The glove is called P5 and is made by Essential Reality.
Courtesy of Essential Reality
This glove was designed for gaming and features flex sensors in the fingers and an infrared receptor that senses the movement of the glove. The device plugs into a USB port and has software to assist in programming it for your application. Best of all we were able to find an online retailer selling them for approximately $30 with shipping and handling. If we are unable to program this glove to control all the movements of the robot we will at least be able to use it for our glove to sense the finger motion.
2 Arm Types
1 Wrist, Elbow, & Shoulder Joints
Our first concept for the robotic arm was made up of the major joints in the human arm. There would be a shoulder joint that would have three degrees of freedom in the x and y directions and a rotation. Further down would be the elbow joint that would have one degree of freedom and then the wrist joint with two degrees of freedom to move up and down as well as rotate. This design would give us motion that is most like the human arm but the problem is that there are too many degrees of freedom to control with our controller designs. We believed that trying to incorporate all these degrees of freedom would make the robot to difficult to operate for the target age group.
2 Wrist, Elbow & Rotating Cylinder Joints
Our second design basically eliminated the shoulder joint except for the rotation part. This design utilizes a pneumatic or electric cylinder to give the arm reach. The cylinder would be mounted so that it could be rotated enabling the arm to reach a circular area. At the end of the cylinder would be the elbow joint that would be able to move up and down like a human elbow. At this elbow joint we are also looking into the possibility of adding a rotation to increase the agility of the arm and hand. It is our belief that by moving this rotating joint from the wrist up to the elbow would reduce the problem of twisting wires and cables that run from the forearm to the hand. The wrist joint would be able to move up and down only.
3 Source of Mechanical Power
1 Pneumatics
Air muscles produced by the Shadow Robot Company could be utilized to power the fingers and various joints in the arm. One of the great benefits of air muscles is they have a power-to-weight ratio as high as 400:1 where electric motors only go as high as about 16:1. The air muscles can apply a force of up to 140 lbs each and are flexible enough to be bent around corners or twisted axially. The general make-up of an air muscle is a rubber bladder covered by plastic woven sheath as shown below.
Photo courtesy of Shadow Robot Company
The normal operating pressure of the air muscles is 0-60 psi. The air muscles would flex and bulge similar to the human muscle creating a more life like robot. In addition to the air muscles one of our concepts involves a pneumatic cylinder for the bicep area of the arm that would allow the arms reach to be increased or decreased as desired. A pneumatic system such as this will incorporate various regulators, manifolds and valves to control the flow to the components. The team was concerned with the noise issues associated with utilizing an air compressor if we would have to enclose it within our display. We have however worked it out with the museum to have them supply an air compressor located away from the display and have the air supply piped to the display.
2 Electric Motors
Controlling the joints by electric motors is another option our group has investigated. There are two main types of motors that could be employed in our project. The first are stepper motors. Stepper motors can be controlled by turning coils on and off; therefore they are easy to control using digital computers. The computer simply energizes the coils in a certain pattern and the motor will move accordingly. An encoder could be attached to the motor to verify its position. The second type of motor is servos. Servo motors are widely used in RC applications like cars and planes. They are a very good match for a robotics project like ours because they are available in a large range of torques and they are relatively small. Inside the small case is a system consisting of a motor, gearbox, feedback device, servo control circuitry, and drive circuit. These motors are extremely easy to control with a digital controller and require about 5-6 volts and draw 100-500ma depending on size.
3 Hybrid System
The third and probably most likely system is a hybrid or mix of pneumatic and electric devices. It is our belief that the air muscles are the best method for actuating the fingers and the wrist joint but the elbow flex and shoulder rotation would be best controlled using servo motors because of their more precise control and positioning. The cylinder could be either pneumatic or electric, at the present time the team is leaning towards electric simply because it would be easier to control the amount of advancement with a worm drive type cylinder for instance. Another reason to use an electric cylinder over a pneumatic one is the amount of air supply needed would be much less since the cylinder would run around 100 psi. Further research must be done into the availability and cost of an electric cylinder before a definite decision is made.
4 Electronic Data and Control System
The electrical control system to be used for this project presented several possibilities which were based on both the controller data input and input and output lines to and from the robotic arm. The possible components were a personal computer (PC) and a microcontroller development board.
1 PC with an output board
The option to use a PC without a microcontroller simplifies the design and the programming necessary. However, this option would require that the PC drive the many input and output lines needed to control the robot. The number of control lines would be between twenty and forty, depending on design decisions such as which kind of valves are used and whether air muscles are used at the elbow joint instead of a servo. For this option there is the issue of sending these lines over long distances and the potential communications trouble that could result.
2 Microcontroller Development Board Alone
The microcontroller development board (MDB) without a PC would certainly consolidate the electrical design to one device and lower the necessary budget. The issue that arises with this approach is that the USB drivers necessary to read the signals from the glove controller would need to be written specifically for the selected microcontroller. This would be a lengthy software undertaking and would require that much of the work and testing of the drivers be redone when the drivers are already written and tested for a PC with the Windows OS.
3 PC with a Microcontroller Development Board
The use of a PC in the design will not be the least expensive method of implementation but it would solve problems with device drivers, required inputs and outputs to the robotic arm, and lengthy communication lines. The PC in this situation would receive the USB signals from the glove controller and a program running on the PC would convert these to an RS-232 serial output from the PC. An RS-232 cable would then take this serial data to the microcontroller which would be located close to the robotic arm. The microcontroller would then convert the serial input to a set of output signals on the many output lines.
Feasibility
Feasibility assessments were conducted to help the team in their selection of the project, controllers, and arm types. We utilized a weighted measurement system to ensure that our key objectives would be met properly.
4.1 Project Feasibility
We used a feasibility chart to prove that the dexterous hand was the right selection for our project. In the chart below the Dexterous hand is set as the baseline project and then the remaining 6 projects are scored on a scale of 1 to 5 on wether they are better or worse than the dexterous hand. The chart show that the hand is the best choice at 100% followed by the labyrinth at 89.3%.
|Dexterous Hand Is the Baseline Project 1 = much worse than baseline concept |Dexterous Hand |Labyrinth |Basketball|Arm |
|2 = worse than baseline 3 = same as baseline 4 = better than baseline 5= much | | |Shooter |Wrestlin|
|better than baseline | | | |g |
|Sufficient Student Skills? |3.0 |2 | |21% |
|Sufficient Financial Resources? |3.0 |5 | |14% |
|Sufficient Time to Complete? |3.0 |3 | |14% |
|Cost of Materials? |3.0 |5 | |7% |
|Has "Attractor Factor" |3.0 |4 | |24% |
|Interactive For Guests? |3.0 |3 | |10% |
|Does Not Require Much Instruction? |3.0 |4 | |7% |
|Promotes Scientific Learning? |3.0 |4 | |3% |
| | | | | |
|Weighted Score |3.0 |3.6 | | |
| | | | | |
|Normalized Score |84.5% |100.0% | | |
2 Arm Feasibility
The arm concepts were placed into the feasibility chart and as shown below the wrist, elbow and rotating cylinder concept is the best design.
|Wrist, Elbow & Rotating Cylinder is the Baseline 1 = |Wrist Elbow & |Wrist Elbow & | |Relative |
|much worse than baseline concept 2 = worse than baseline 3 = same as |Rotating Cylinder |Shoulder | |Weight |
|baseline 4 = better than baseline 5= much better than baseline | | | | |
|Sufficient Team Skills? |3.0 |2 | |21% |
|Sufficient Financial Resources? |3.0 |2 | |14% |
|Sufficient Time to Complete? |3.0 |2 | |14% |
|Cost of Materials? |3.0 |1 | |7% |
|Has "Attractor Factor" |3.0 |3 | |24% |
|Interactive For Guests? |3.0 |3 | |10% |
|Does Not Require Much Instruction? |3.0 |3 | |7% |
|Promotes Scientific Learning? |3.0 |2 | |3% |
| | | | | |
|Weighted Score |3.0 |2.3 | | |
|Normalized Score |100.0% |78.2% | | |
3 Electronic Control Feasibility
The electronic system can involve either a Microcontroller with a PC, or a PC with an output board, or a Microcontroller only. The aspects of choosing each are shown in the feasibility assessment below. The team is in favor of the Microcontroller with PC because of the existing USB drivers and better communication properties. The major drawback is the cost and complexity of design but the savings is in software development time.
[pic]
5.0 OBJECTIVES & SPECIFICATIONS
A set a guidelines, design objectives, and performance specifications was established to assist the team in properly assessing on how successful the outcome of the project is. The following sections of the chapter will go through the different objectives, specifications, and guidelines that the team agreed upon.
1 Design Objectives
There are a number of design objectives that required the attention of the team. These objectives have to be specified in order for the team to have a list of goals and aims to achieve. These objectives are listed below:
1. Provide a robotics display with a strong “attractor factor.”
2. Provide a display that will stimulate a child’s interest in learning about robotics and engineering.
3. Produce aforementioned display at a cost less than $5,000.
4. Enhance the learning experience at the RMSC for all that visit the facilities.
5. Design a robot that can perform tasks that the normal human being may not be able to do (palm a basketball).
2 Performance Specifications
It is inevitable that as the project continues, the team will face numerous obstacles and problems. However due to time constraints, not every issue will be addressed. By having a list of performance specifications, it will aid the team in prioritizing what is crucial. This will help manage time more wisely into what problems must be fixed and which obstacles the team can overlook.
1) Robotic Display Shall be intuitive and require little to no instruction prior to operation.
2) Robotic Display Shall be easily maintained and repaired.
3) Robotic Display Shall have a strong “attractor factor,” drawing children to want to come play with it.
4) Robotic Display Shall rest on a table which is 28” in height.
5) Robotic Display Shall fit through a 4’ door opening.
6) Robotic Display Shall be moveable by two people.
3 Design Practices
To help the team achieve the objectives and specifications that was established, a list of design practices were kept in mind when team members were developing designs. A list of these practices is as follows:
1. Design for Manufacturability and Minimum Cost – When possible, we tried to purchase parts off the shelf for this project. While designing the custom components of the robot we took care to design it so that it can be most easily manufactured. We did this by simplifying our design as much as possible and calling for standard sizes and materials whenever we could. We kept tolerances as loose as possible and made custom parts easily machinable.
2. Design for Assembly - We produced this display so that each component can be built separately. That way each of us can focus on a separate part, making the group more efficient. However, when the parts are brought together they can be easily assembled.
3. Design for Optimum Alignment
4 Safety Issues
To ensure the safety of all members on the team, a set of safety precautions were established. Since the testing of the design will undergo high pressures and components will be spinning at high speeds, it is imperative that the members of the team follow these guidelines.
1) Since this display will be targeting a young audience we designed it to be enclosed, thus eliminating most safety issues.
2) All proper precautions will be made in order to safely run air and electricity through the system.
3) Health concerns from the repeated use of a glove were the only other issue that arose. We chose an open glove which eliminated this problem.
8 DESIGN ANALYSIS & SYNTHESIS
1 Display Analysis & Synthesis
The Display unit for the Robotic Arm is designed according to RMSC specifications. The base of the unit is 28” above the ground, which makes the display low enough for children and the handicapped. It has been built out of steel and painted aluminum color. We planned most of our calculations around that specification. It also fits through a 4ft door opening so that it can be easily transported to the museum. The display is approximately 6ft tall in order to achieve our goal of having a robotic arm that is twice the size of a human arm. The sides of the display will be of a glass/plastic nature and will be taken care of by the RSMC display builders. They will be made from a material that will not break or shatter on impact. The display will be used by many children and will have a design such that it cannot be broken easily.
[pic]
2 Hand Design Analysis and Synthesis
The hand is designed to be twice the size of a human hand and is made out of Aluminum. The hand has several different components, which include a palm, 5 fingers (four fingers with three “bones” and one thumb with two “bones,” and a wrist joint). Each portion of the hand is connected together using pins purchased from McMaster-Carr. These pins allow for free motion of the joints, which mimics human movement.
The four main fingers are comprised of three different sections each of equation dimension. The dimensions of each section are approximately 2” x .75” x .5”. The thumb will have the same dimensions but is only sectioned into two “bones.” The maximum pressure allowed on each finger is limited to approximately 15lb (Fmax=15lb). Using simple moment equations such as M=Fd we found the maximum moment in the fingers to equal 75lb-in. This moment can be found at the base of the fingers where they connect to the palm.
While this is a maximum moment, it is not large enough to cause fracture in the connecting pin. The pins used have a Shear Strength of 5,400 lbs/in. We are not worried about the aluminum fingers fracturing due to their size and the minimal loads applied.
The palm is also made using aluminum. It is designed to have the same hinged joints with pins to connect each of the fingers. It also has a joint on the opposite end of the fingers to act as an elbow. The palm is hollow to allow for a smooth passageway for the strings or wires that will control the movement of the fingers. The strings stretch from the forearm to the palm through holes drilled into the back of the palm. The string then goes through the finders by small holes cut into the joints. This string then pulls on the finger in order to bend each joint.
The hinge on the back of the palm for the wrist movement has been analyzed for several different joint movements. In order to perform calculations on this joint we first needed to find the weight that it would support. This joint will support 10lb at most. This weight includes the weight of the aluminum hand plus the object that the hand is lifting. The lifting action of the hand will create a moment at the wrist. The maximum calculated moment at this joint was found to be 50lb-in which can be seen when it is at a 90 degree angle with the vertical axis. Again, this moment will not cause a large enough stress to create a fracture in the joint and/or pin. The hand is shown below. Each individual portion of the finger is also displayed.
[pic]
[pic][pic]
Palm Finger Tip
[pic]
Finger Joint
6.3 Forearm Design Analysis and Synthesis
The forearm is a very simple design. It is a galvanized steel pipe that was purchased from Home Depot. It is 1.5” in diameter and approximately 15” in length. The pipe is great for our design because it allowed us to drill holes in it to attach the upper arm cylinder, the palm, and several eyehooks for the air muscle attachments.
Again, this portion of the assembly will not fail because of the strength of the galvanized steel. Steel is used in many applications and can bear an extremely high load. The stress that can be expected on the forearm is approximately 6 psi while the yield of the steel is 50-75 ksi. Clearly, this will not incur any failure.
6.4 Upper Arm Design Analysis and Synthesis
The upper arm is an electric cylinder purchased from ebay. It was originally a product from IDC Motion. The base is approximately 20” in length and has a cylinder that protrudes an additional 18” upon activation.
The upper arm is connected to the top of the display using a mounting bracket and a Lazy Suzan. This allows the cylinder to rotate about the vertical axis, much like the human shoulder joint. This is the only movement of the cylinder besides the actual in/out movement of the rod inside the cylinder. The cylinder is forced to rotate by user inputs that control a motor. The motor turn a belt that drives the Lazy Suzan to turn.
[pic] [pic] [pic]
[pic]
Material Design.
Our project consists of many different materials. For our machined parts we chose aluminum for its ease of machining and relatively low cost in comparison to other options such as composite. As an issue of cost, we also machined a steel tube to make the forearm. For strength we chose steel for our display case. We also used wood in some of the parts we manufactured. The display contains a 2 X 4 as a mounting bracket and plywood as the base of the display. Everything is to be painted for aesthetic purposes prior to the display being put on the museum floor.
Parts.
Our project consists of many different parts, both purchased and machined. On the purchased side are air muscles which came from a specialty company in London called the Shadow Robot Company, Valves, regulators and other pneumatic accessories came from a local company called Numatics, we purchased an electric cylinder with a 19.5 in (495 mm) stroke made by IDC Motion which was used, saving us substantially on cost, two servo motors which were purchased as used parts, and an aluminum turntable, or “Lazy Susan Bearing” which came from McMaster-Carr. We also purchased torsional springs, pins, and clevis joints from McMaster - Carr. The necessary steel and aluminum came from a place in Rochester called the Steel Supermarket on Mt. Read Blvd.
As far as machining is concerned, we machined the hand, fingers, forearm, top plate, frame, and several other specialty parts. The hand was machined from a solid block of aluminum using the CNC lathes on RIT. We then had to drill many holes into it, fit a clevis-type joint to it and put a Plexiglas cover on the bottom. The fingers were machined from bar stock and milling and drilling were required to form them how we needed. We also had to fit torsional springs between the joints to cause the finger to return to its original position after being bent. Homemade joints hold the fingers to the hand and all joints are pinned together to allow them to curl smoothly.
The forearm is made from a piece of steel tubing; we machined joints into the end so that they could be pinned on both sides. We also had to mill a clearance gap in the upper side to allow rotation of the elbow joint. Furthermore, the tube has holes drilled and threaded rods inserting to allow us to mount the air muscles. The tube is attached on the upper end by a clevis joint which screws on to the electric cylinder. The clevis joint is also machined. We built a pulley into it and also drilled a hole into it to allow function of the air muscles.
Next is the electric cylinder. First of all, the cylinder contains a mount for the motor which we machined. This was necessary because we replaced the stepper motor it came with with a servo motor which we purchased. The cylinder is held in place by an aluminum plate we machined. The plate has holes drilled into it and screws go straight down into the electric cylinder, holding it in place.
The plate is drilled into a “Lazy Susan Bearing” which is mounted on top of 2 X 4’s, which are in turn mounted to the display case. On top of our mounting plate we secured a timing pulley which works in conjunction with a servo motor. The servo motor hangs upside down from the display by a custom mount which we machined ourselves.
Operation.
There are many aspects to the mechanical operation of this project. The entire arm assembly has the ability to rotate. This is achieved by a servo motor, which is mounted to the top of the display case. The motor turns a timing belt, which is attached to the mounting plate, which is attached to the “Lazy Susan Bearing.” By turning the motor we can rotate the electric cylinder and therefore, the entire system. We have allowed for the arm to rotate 180 degrees. This allows access everything in the display, without making the display so big that it cannot fit through the museum doors.
For extension of the arm, the electric cylinder contains it’s own motor. That motor turns a worm gear within the cylinder and causes it to extend. This particular cylinder allows for 19.5 in (495 mm) of extension. In our display case, this allows the hand to reach to the floor when the hand and forearm are parallel to the floor of the display.
All joints (fingers, wrist and elbow) are controlled by either one or two air muscles. The muscles are pneumatic and linked to our valves, flow controls, regulator, filter etc. One valve controls each finger joint, with the exception of the middle and ring finger. The middle and ring finger are controlled by a single valve and thus, their motion will be tied together. This was a precaution made so that kids at the museum will not be able to use the hand to give the middle finger. The wrist has an air muscle on both the top and bottom, both of which are supplied by the same valve. This is a different valve, however and as one side charges, the other discharges, allowing the wrist to flex in both directions.
We have two sizes of air muscles, the small, which produces 6 lb (27N) of force and the large, which produce 40 lb (178N) of force. The air muscles are attached to the forearm. One side of the muscle is tied off and cannot move. A string runs from the other end of the muscle to what it moves. So the small muscles run through the finger to the fingertips and the large muscles run to a place just below or above either the wrist or elbow joint. The muscles’ natural positions are in tension, however, when they are inflated with 60 psi (4 atm) of pressure, they contract, working much like “finger handcuffs.” The pressure causes the muscles to contract, pulling the string on the non-stationary end. That string then pulls the fingers or other joints and causes them to bend. By pulsating the valves, you can precisely control the position of the fingers, elbow or wrist. The returning force is provided by torsional springs in the fingers, a second air muscle in the wrist and merely by gravity in the elbow.
The entire mechanical system is linked to the electrical system through a microcontroller and a PC.
Lifespan.
The pneumatic valves are the only component that provided a tested lifespan. The Numatics Incorporated rates the valves we purchased for 200 million cycles. The museum gave us data that the museum operates 8 hours a day 363 days a year. In the tables section of this picture is a chart showing the projected lifespan of the valves for various cycles per minute assuming display will be under constant use for the full 8 hours a day. We are estimating the average use to be 70 cycles per minute which would equate to a lifespan of 16.4 years.
[pic]
Figure 7: Lifespan of Air Muscles [4]
Stresses.
We designed our entire assembly so that there would not be any structural failure due to stresses. However, stresses are still a consideration in any design project. The area of most concern in the arm assembly is the elbow joint. This is the joint that will encounter the most stress because it will be supporting approximately 15 lbs of material. One portion of the joint that could encounter relatively high stress concentrations is the clevis joint and the connector pin. The connector pin and the clevis joint each have a shear strength of 5,400 lbs (McMaster-Carr).
There can be a calculation for the axial stress that will be felt by the pneumatic cylinder. With the cylinder bearing a maximum force of 15 lbs with an area of 1.327 square inches we can calculate the axial stress. This stress is equivalent to the force divided by the area. The axial stress on the cylinder is found to be 11.3 psi. The Ultimate Tensile Strength of this steel cylinder is 74.5 ksi. Clearly, this cylinder will not fail under axial loading.
The forearm can also have a stress calculation completed on it. The shear stress on the forearm can be found using the following formula: [pic]
Using this formula we calculate the stress to be 5.97 psi. Given the Yield Strength for the galvanized steel pipe, 50ksi, we get a factor of safety much greater than one.
The stress throughout the assembly will never exceed the maximum allowable stresses for the materials given. After many calculations we can see that our assembly will not fail as a result of stress failure.
6.5 Equations Used for Mechanical Design Analysis and Synthesis
[pic]
[pic]
11 Electrical Control
The electrical control consists of three main parts, the personal computer, the microcontroller, and the output power circuit.
Design Overview. The electrical design problem consisted of designing an interface that detects the movements of the human hand and translates them into control signals for the robotic arm. The user input device was chosen for its apparent usefulness for our project, for its ease of replacement, and for its low cost. It was decided early on that there might be a need for many input and output signals because of the many mechanical devices that needed to be connected. This led to a possible design including a microcontroller with many input/output pins or a special input/output card attached to a personal computer. The end design decision to allow for the greatest flexibility includes both a personal computer and a microcontroller. The microcontroller does not supply the necessary voltages and currents to drive the mechanical and pneumatic devices so a complex circuit design was implemented involving power supplies, relays, and transistors.
User Input. In order to best accomplish a feasible solution in terms of time and money, the input device was purchased commercially. The P5 Glove Controller includes five bend sensors, one per finger, mounted on a sturdy plastic frame. Each sensor provides a 0.5 degree resolution independent finger measurement. In addition to the unique finger bend measurements, the glove provides detailed position tracking of six different positional axes through the use of a remote infrared sensor. The glove worn on the right hand of the user has eight infrared lights on its plastic shell. The infrared sensor tracks six axes which include x, y, z, yaw, pitch, and roll.
Personal Computer Software Design
The PC design included the use of the drivers and free software development kit supplied by the manufacturer of the P5 glove. The software also included the use of serial communication via an RS-232 serial link with the microcontroller. The data coming from the P5 glove was made available through the provided functions of the software development kit and then converted into 8-bit bytes for transmission over the serial link.
[pic]
Figure 2: Simple Diagram of the PC software
The P5 data was in terms of absolute position with respect to the glove’s starting position and thus was not useful in terms of absolute positional control. The output data would wander and never return to a reference location. Thus the decision was made to simply use the change in position (velocity) to control the robotic arm. When a certain threshold voltage is reached, the program responds by enabling the motion of the corresponding robotic arm component. The arm components with their corresponding control signals from the glove are given in Table 2.1.
Microcontroller Design.
The function of the microcontroller is primarily to take the serial data in and to send the digital output signals to the individual control lines for each robotic component. The output control lines are given in Table 2.2.
Table 2: Microcontroller Outputs
The microcontroller operates at 3.3 digital high voltage and has a limit of 25 mA on the outputs of the ports. The microcontroller chosen for the design is the C8051F020 which contains multiple analog to digital converters, 64KB of programmable FLASH memory, 2 UART serial interfaces, five general purpose 16-bit timers, and a programmable counter/timer array.
The serial communication is performed via the UART0 with a set baud rate at both the PC and the 8051 of 4800 bps. This baud rate is slow compared to the capability of the system but there is no need for higher resolution because of the high frequency filtering of the input signals in the PC software.
Circuit Design
First we use a 3V power supply and NPN Bipolar Junction Transistor to amplify the signal output from the microcontroller. For the amplifier we chose the 2N3904 shown below in Figure 3.
[pic]
Figure 3: BJT used for Output Current [2]
The 2N3904 has:
IC=200 mA IC = 0.1 mA, VCE = 1.0 V
Each output will control one relay. [pic]
Figure 4: Low signal relays G6K-2P-Y [3]
All relays will get the amplified signal from the microcontroller. Relays will perform a simple on and off function to supply 24V power.
We chose Omron’s S8TS 24V power supply pictured in Figure 5.
[pic]
Figure 5: Project Power Supply [3]
One power supply will only connect to the cylinder motor which allows the hand to elevate up and down. The motor is 24V 1 Amp so that the reason for having one power supply for this motor is that this DC motor has a very high start up current. When the motor is on the power supply will drop so a large amount of current is drawn and the voltage drops down. After a short interval of approximately a tenth of a second, the motor will return to the desired operational current draw 0.98 Amp.
The additional power supply will serve all the other power on the board.
Considering the high start current through the cylinder motor, the power supply will be hot. We added a fan on the side of the power supply. The fan is 12V 0.08Amp.
The LED is the output from the relays output(24V) which is for simulation and program debugging.
[pic]
Figure 6: Simple Circuit Block Diagram
6.6.1 Electrical Input and Output
The input to the PC is the P5 Glove controller. It sends data via its USB cable. The PC output is the RS-232 serial link which is attached to the PC’s com port. The control inputs and outputs from the robotic arm and hand go to output-configured ports of the microcontroller. The input-configured ports receive the limiting sensors for the extension cylinder, arm rotation, and floor pressure.
13 Electrical Control Algorithm
The control algorithm is quite simple and depends only on the motion and not position of the glove controller. When the glove controller is moved, the robotic arm will also move. The position of the arm will be adjusted by the person who is controlling the arm.
The arm moves its individual components based on these signals from the glove:
‘x’ Axis controls rotional motor
‘y’ Axis controls elbow motion
‘z’ Axis controls extension cylinder
‘pitch’ Axis controls the wrist motion
Finger Motion Control:
Motion control theory for the fingers is same as for the arm
Each robotic finger is controlled by each finger sensor on the glove
6.6.3 PC Program in C++
The C++ program written for the PC performs 3 continuous operations. It reads the new user input data from the glove, filters and translates it into output serial communication bytes, and sends it out to the serial port.
• Initialize the P5 Glove
• Initialize the Serial Port
• Begin Main Loop
• Update Glove Variables with current data
• Filter Incoming Data to detect motion above a certain threshold
• Convert Filtered Data to an 8-bit serial output
• Send output byte to the serial port
• Loop (forever)
6.6.4 Microcontroller Program for 8051 MCU
The assembly language program written for the 8051 performs 3 continuous operations. The main loop polls the serial port flags to see if new data has arrived. When a new byte arrives the first 3 bits are processed to determine which motion should be acted upon. The input from the limiting sensors is checked to make sure this motion is allowed and then the corresponding port pins for this motion device are updated.
• Setup the UART for serial transmission
• Setup the port pins for sensor inputs and signal outputs
• Begin Main Loop
• Look at the most recent incoming serial data
• Check the first 3 bits for motion ID
• Check the last 2 bits for desired motion
• If the sensor feedback allows this motion, execute this motion and send output to port
• Loop forever
7.0 DELIVERABLES
The important deliverable is a functional display to be delivered to the RMSC, but there are a few additional deliverables due to the nature of the project and the requirements of the RIT Senior Design class. These deliverables are to be turned in to the RMSC and thus do include the deliverables for RIT Senior Design.
7.1 The Robotic Arm and Hand Display
The functional display deliverables include the P5 Glove, the personal computer with functioning glove code, the USB connection to the 8051 microcontroller, the fully programmed 8051 microcontroller with port connections to power transfer circuit, the power transfer circuitry connected to the motors and valves, the display case with the mounted arm and hand.
7.2 The Robotic Arm and Hand User’s Guide
The user’s guide for this project is necessary because the project will be delivered to the RMSC upon completion of Senior Design by the project team members. The user’s guide will include information concerning the normal operation of the mechanical and electrical components. It will also provide a resource for maintenance and troubleshooting, should the arm cease to function normally.
7.3 The Complete Parts List with Vendors
The parts list will include all the parts used in this project and will provide replacement parts for each with the vendor name and price.
8.0 PROJECT TIMELINE
The following is the timeline for our Senior Design Project. There was a need to have additional time for design which is shown during the first part of the Winter Quarter because of the few short weeks available before PDR that the team had for design. The team was not able to test the project at the RMSC until after CDR due to the scope and difficulty of the project.
Senior Design Fall Quarter:
Senior Design Winter Quarter:
PROJECT BUDGET
The estimated cost on some items corresponds to the actual component price from a known vendor. However other items either need to be chosen for the design, or a vendor may still need to be chosen. Further design analysis needs to be done on items such as the microcontroller and valve type and the budget will change if some components are donated.
|part |Quantity |Unit cost |Estmated Cost |Actual Cost |
|Sensor Gloves |1 |$80 |$80 |$30 |
|Air Muscles large |5 |$50 |$250 |$250 |
|Air Muscles small |6 |$26 |$26 |$156 |
|Valve Manifold |1 |$160 |$160 |$160 |
|Valves |4 |$70 |$280 |$280 |
|Regulator |2 |$50 |$100 |$100 |
|Air Line |3 |$15 |$45 |$45 |
|High Tensile String |1 |$12 |$12 |$12 |
|Ait Filter |1 |$75 |$75 |$75 |
|Hand and Forearm |1 |$500 |$500 |$383 |
|Electric Cylinder |1 |$200 |$200 |$65 |
|Enclosure/Display |1 |$1,000 |$1,000 |$450 |
|Microcontroller |1 |$200 |$200 |$180 |
|PC |1 |$400 |$400 |$0 |
|Electric circlts |1 |$400 |$400 |$559.09 |
| | | | | |
|Total | |$3,238 |$3,728 |$2,745 |
[pic]
-----------------------
[pic]
Motion of the corresponding robotic fingers
Motion of the wrist
Motion of the cylinder (up/down)
Motion of the elbow
Rotation of the entire arm
楐捴൨复洠瑯潩൮堍洠瑯潩൮䘍杩ㄠㄭ›删扯瑯捩䄠浲愠摮䐠硥整潲獵䠠湡慂楳敓畴൰䘍杩㌠㈭›倠獯楳汢潊獹楴正眠瑩畐桳䈠瑵潴獮映牯䄠摤瑩潩慮潃瑮潲൬名慲獮業⁴慄慴瘠慩匠牥Pitch
Y motion
X motion
Fig 1-1: Robotic Arm and Dexterous Hand Basic Setup
Fig 3-2: Possible Joystick with Push Buttons for Additional Control
Transmit Data via Serial Link
Convert Data to Byte Messages
Z motion
Get current glove data
Set up Serial Comm
Initialize P5
Fig 3-4: An Air Muscle showing how it increases length with applied pressure
Rotate Arm Left
Move Wrist Up
P4.2
P4.3
P4.5
P4.4
P4.6
P4.7
P4.0
P4.1
Extend Arm Up (Cylinder Actuator)
Extend Arm Down (Cylinder Actuator)
Rotate Arm Right
Move Elbow Down
Move Elbow Up
Move Wrist Down
Pinky Finger Contract/Release
Middle OR Ring Finger Contract/Release
Index Finger Contract/Release
Thumb Finger Contract/Release
Output Control
Glove Input
Finger bend
Fig 3-3: Robotic Arm and Dexterous Hand Basic Setup
Fig 3-1: Basic Flex Sensor and Circuit
Micro-controller
P5.1
P5.0
Fig 3-5: Robotic Arm and Dexterous Hand Basic Setup
USB
PC
P5.2
RS-232
P5.3
Output
Port
Robotic Arm and Hand Action
Microcontroller output (3.3V 25mA)
BJT 2N3904 offer more current
Low signal relays G6K-2P-Y
Power supply
Motors and air muscles
-----------------------
Robotic Arm & Dexterous Hand Senior Design 05911 04013
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.