Project Proposal (11)

Justin Hung, Mateo Oviedo-Cardona, Aakarsh Saggar, Yaman Abdulkader

Robot Mechanism Design (8-9:30AM) 10/14/2025

Introduction

Robot arms frequently have motors mounted directly at each joint to drive successive arm segments. This results in excess weight which creates a constant load on the arm’s joints and adds inertia. This reduces its effective load capacity, and slows down its motions.

Problem Statement

“To design and construct a proof-of-concept mechanism that eliminates the need to mount motors on the arm itself, increasing the load capacity and speed of movement” The main challenge is in achieving a variable motion profile as measured from the end of the arm, using motors not mounted to the arm itself. Simple joints are only capable of repeating a limited range of motions which is not acceptable for a robotic arm. The motion profile must encompass a full hemisphere centered on the arm’s base. Complexities arise when attempting to keep the mechanism compact and light enough to provide a marked improvement over the original solution (motors on arm). The motion of the arm segments relative to each other introduces significant challenges as there must be a connection between segments in order to transmit motion to the wrist at the end of the arm.

Mechanism

The primary mechanism used in the solution will likely be the differential as it allows for the transmittance of various motion profiles through a rotating joint. Each joint will have a series of differentials to send power up to the next segment of the arm. Instead of traditional links and joints, the gears and their meshing will create their equivalents. This combination of idealized links will move in 3 dimensions since the gears rotate and are oriented perpendicular to each other.

Open

Fig. 1: Diagram of a differential using bevel gears

Open

Fig. 2: The simplified diagram of a set of bevel gears

Proposed Scope

The team will at a minimum, develop a theoretical analysis of the motion profile, a model of the arm’s motion for different motor inputs, and a proof-of-concept model. The design only involves that of the arm itself. No end effector is included due to the lack of a one-size-fits-all effector as well as the added complexity of accommodating one. However, a rotating wrist will be included. The proof-of-concept model may not be the robot arm itself. For example, if the differentials required to transmit power through the joints are too difficult to manufacture, the model may simply use separate motors to simulate the motion inputs from one segment to the next.

Open

Fig. 3: Inputs into the first arm segment result in an output motion. The output motion is then simulated as a series of inputs into the second segment, forgoing the need for physical power transfer through the arm joint.