Robotic Arm
The Project
Designed and built a controllable robot arm. The robot has six degrees of freedom and can reach nearly all objects within 18 inches of the center of the arm. Though similar robots already exist, this project pushed me to make design decisions and have them realized. Additionally, I solidified my skills with useful tools such as CAD software, microcontrollers, laser cutters, and more.
Research
Using the internet as a resource, I found several examples of similar projects. By reading blogs and watching videos, I learned some key challenges of such a project. Some things I observed from other projects are:
Custom shapes for the arm can be useful, but expensive to manufacture
An arm with many joints requires a large amount of torque at the joints
Controlling the robot can be difficult to operate
Image from ‘howtomechatronics.com’, an example of a similar project I analyzed
By learning from other people, I was able to reduce the amount of work required for my project, as well as improve its performance. Research on the current market is a critical part of designing a great product. By looking at several examples, I could foresee some potential pitfalls and avoid them entirely. I believe that this is an important step to perform before embarking on a design project.
Early stage sketch
The Design
Using the research I had done on similar projects, I made educated decisions before designing the robot. Some examples of design decisions I made are:
The structure of the robot would be made of a material with uniform thickness to simplify manufacturing and reduce cost.
The motors would be high-strength servo motors to avoid motor failures; which I saw in other robots.
The claw would utilize a 4-bar parallel guided mechanism to improve grabbing abilities.
Design Successes
These design decisions helped immensely with the simplicity of this project. When I first began the project, I envisioned a highly complex system where I printed many parts and the arm worked effortlessly. The research I did helped me to understand some realistic expectations for the project. Then the design decisions I made helped me to make the project a reality. For example, using a uniformly thick structure allowed me to laser cut the parts from the same material. This process was simpler, cheaper, and faster than my original vision of printing individual parts. On a similar note, choosing high strength servo motors from the start saved me money and time I could have wasted with cheaper motors. A more detailed image of the design can be seen in the CAD model shown below.
Design Weaknesses
Despite many good design decisions, my robot design had many weaknesses. First, the arm was heavy. I didn’t account for the strength needed to lift such a heavy part. Although I knew that I needed strong motors, I didn’t take the effort to reduce the weight of the robot. Such a change would have been easy by stripping the wooden parts of the unnecessary mass. To account for this, I had to add springs into the design to make up for the robot’s weakness. Second, the controls were attached to the base of the robot. This limited the movement of the arm as it couldn’t reach the area where the user was positioned. A remote controller or pre-programmed path could be used to overcome this weakness. Finally, the claw had insufficient strength to lift many objects. Since the joints were rods connecting two sheets of wood, the claw had minimal stability laterally. This flaw greatly limited the robot’s capacity to move objects. The gravity of this flaw may merit a complete redesign of the claw mechanism. These weaknesses are a few of the obvious ones I observed from an initial prototype of my design, and I recognize there were more limitations not mentioned.
Building the Prototype
To build the arm, I needed to assemble all the parts I included into the design. For this project I only created one prototype. If I was creating a good product and was given ample time, I would have iterated through prototyping many times rather than just one. The prototype I created consisted of many out-sourced parts such as the servo motors, Arduino board, and Lazy Suzan. Custom parts were also included and created using 1/4” medium-density fiberboard (MDF) board.
CAD model
I created a computer aided design (CAD) assembly of the structural components of the arm I designed using SOLIDWORKS2016 software. This activity both helped in design decisions (such as the length of members) and served as a template for laser cutting parts later. I have frequently used CAD software to generate geometry needed for assembling prototypes, manufacturing parts, and finite element analysis. A rendering of the CAD assembly is shown in the image above.
Robot Construction
Constructing the robot consisted of attaching all the parts included in the design. The body of the arm was cut using a laser cutter patterned after the CAD model I created earlier. Motors, joints, springs etc. were assembled by hand with specified screws, fasteners, and glue. The electronics were hidden in wooden base of the robot. Motor connections ran through the body of the robot down to the controls.
Controls
An integral piece to this project was the ability to control the arm. The controls consisted of two joysticks and a button. The joysticks controlled five of the servo motors for arm mobility, and the button controlled the claw. Arduino was used as the brain of the robot. Using the joysticks to move the robot resulted in constant speed sweep of each servo motor. The speed of movement was made to be relatively slow so that the arm was easier to control.