Autonomous Motion Planning for a Motorized Walker Using Potential Field and Admittance Control

Assistive walkers are essential for many individuals who require geriatric care or mobility aids for day-today activities. The functionality of these devices is not able to provide the care that is needed for those with different levels of physical disabilities or who are visually or mentally impaired. Developing an autonomous powered walker that can receive a user’s intent and assist them in navigating obstacles would allow more people to have access to enhanced mobility. In this work, a control strategy is designed for obstacle avoidance using potential field generation and admittance regulation. The mechatronics hardware and software are developed to generate and implement a continuous real-time motion trajectory for the smart walker in response to the user’s interaction torque and scanned locations of obstacles in the environment. The integration of the mecanum wheels, DC motors, microcontrollers, a rotating LiDAR scanner, and a mini-PC powered by an onboard battery allowed for the evaluation and testing of the obstacle avoidance algorithm and admittance control in real time. The experimental results showed that the autonomous walker can generate desired trajectories in response to the user’s interaction using the admittance controller and navigate around the obstacles with real-time velocity updates for the motorized wheels.