Atif Mahmood

Gait trainers are increasingly used in rehabilitation for patients with lower extremity impairments; however, current control strategies often neglect the inertial forces from the body’s vertical motion, limiting effective force compensation. This results in deviations in the unloading force of rehabilitation devices, disrupting the patient’s natural gait. We introduce an estimation technique to predict vertical inertial forces generated by body movement, enabling rehabilitation robots to actively compensate for inertia during therapy. Our approach utilizes widely available sensors, such as smartphones, providing a cost-effective alternative to traditional multi-sensor setups, reducing wiring, and minimizing movement restrictions, thus enhancing data collection and preserving natural gait. Additionally, this technique has potential applications for Low Gravity training in NASA’s Active Response Gravity Offload System (ARGOS), supporting astronaut training by simulating reduced gravity conditions.