Simulation and Control of an SO-101 Robotic Arm for Assistive Daily Living Tasks

Assistive robotic systems have the potential to improve independence and quality of life for individuals with mobility impairments by helping them perform everyday tasks. In this article, I investigate the feasibility of using low-cost robotic arms to assist wheelchair users with object manipulation and retrieval. I first assembled and calibrated a physical SO-101 robotic arm system consisting of a leader and follower arm, each with six degrees of freedom. The follower arm was configured to replicate the leader's movements through teleoperation. I then implemented the SO-101 arm in the MuJoCo physics simulation environment and trained it to perform object manipulation tasks, including reaching, grasping, and lifting objects. Building on this work, I developed a simulated assistive wheelchair equipped with two SO-101 robotic arms mounted beside the armrests, providing a framework for exploring coordinated robotic assistance. The proposed system could eventually help wheelchair users retrieve dropped objects, pick up drinking bottles, handle medication containers, and access everyday items without caregiver assistance. This article establishes an initial framework for integrating affordable robotic manipulators with mobility devices, with future work focusing on coordinated dual-arm manipulation, autonomous object detection, and real-world implementation.