Gesture Controlled 6-DOF Robotic Arm

Senior Design Project, SUNY University at Buffalo (B.S. Mechanical Engineering)

Thiha Zaw, Thaw Dar San, Katelyn Churakos, MD Hossain

Overall CAD Design

Overall Design

Abstract

This senior design project presents the development of a low-cost ($500), gesture-controlled robotic arm intended to enhance independence for wheelchair users with limited hand strength. The objective is to validate both the sensing-to-actuation pipeline and the structural concept for a compact 6-DOF arm capable of safely lifting 300 g. The prototype employs Leap Motion–based gesture control, an Arduino with CNC shield driving stepper and servo actuators, a 2020 aluminum extrusion frame, and custom 3D-printed joints and gripper. Methods include detailed CAD design, bench-level wiring and integration, constant-increment motor control with endpoint limits, and structural verification using SolidWorks Simulation on the forearm extrusion and a critical printed joint. Interim results demonstrate stable single-DOF closed-loop motion, a fully assembled frame with verified wiring, and total parts cost within the allocated budget from the University at Buffalo Department of Mechanical Engineering. Finite element analysis indicates a minimum safety factor of ≥ 3 for the robotic arm. Future work includes integrating Leap Motion for multi-DOF control, completing FEA-driven design updates, and implementing additional safety features in accordance with relevant personal-care robotics standards.

CAD Design Development

Six Degrees of Freedom was chosen as the final kinematic design for this robot since 6 DOF can achieve any spatial coordinates within the reachable workspace while also maintaining yaw, pitch, roll orientations. The overall SolidWorks development of the product and explanation for the design decisions are as shown in the following figures.

Exploded View of Base Joint

Base Joint

Exploded View of Shoulder Joint

Base and Shoulder Joints

Exploded View of Elbow Joint

Figure 5: Wrist Joints

Figure 6: Parallel Gripper as an End-Effector

Finite Emelemnt Analysis

Figure 7: Side View

A Finite Element Analysis was carried out to determine the rigidity and strength of the 3D printed Components of the robot arm. As shown in Figure 7, the weakest point of the arm is located at the 3D printed "Base Joint" frame, since it needs to carry the load of the entire arm including the payload. Therefore, FEA analysis was done on this "Base Joint" component and the length of the robot arm as well as the payload capacity were determined.

To prepare for FEA analysis, the geometry of the part was first simplified to reduce the computational load. The part has one plane of symmetry in the middle so, it was cut in half and feature of symmetry was applied. The complicated features such as screw holes and counterbores were also removed since they are not playing crucial role in this case. To ensure that there is no singularity, the fillets and round features were also added to the sharp corners.

The Boundary Conditions applied are as follows,

Constraint: One end of the joint with the circular frame will be connected to the shoulder joint. Thus, this portion can be treated as the planar constraint (Roller/Slider) for the boundary condition. Then, bearing constraint features are applied to the four screw holes since screws will be used to secure the connection.

Loading: Force loading acting to the rectangular hole when the robot is fully extended was obtained form the hand calculation. Then, this force was applied to the inner region of the rectangular hole.

After the analysis, the results are obtained as shown below.

Figure 8: Similified Part Geometry

Figure 9: Stress Fringe Plot

Figure 10: Displacement Plot

Manufacturing and Final Assembly

Base Frame

Shoulder Joint Casting

Robot Kinematics, Control, and Programming

MATLAB Simulation

Developed and implemented forward and inverse kinematics for the robotic arm, including a Jacobian-based inverse kinematics algorithm, and validated the control strategy through MATLAB Simulink and Simscape Multibody simulations.

Leap Motion Controller

Integrated Leap Motion hand-gesture control, allowing intuitive real-time human-robot interaction where users command the arm through natural hand movements.

Arduino Code Development

Please see the codes at: https://github.com/ThihaZawWJ/gesture_controlled_robotic_arm.git

Or simply click the GitHub Logo on the right side.

Acknowledgement

We would like to thank our faculty sponsor, Dr. Reza Rashidi, for his continued dedication to the success of this project. Additionally, our teaching assistant, Rishabh Shukla, has provided invaluable feedback that has been integral to our success. The project could not have succeeded without Dr. Rashidi’s and Rishabh’s insightful assistance. We would also like to thank the University at Buffalo’s Mechanical and Aerospace Engineering department for supplying the necessary funding. Without this funding, this project would not have been possible.

Team Photo

Katelyn, Thaw, Me, and MD [from left to right]