Engineering Contributions in the HEPHAISTOS Project

As an engineer in the HEPHAISTOS project, I contribute to the development of an innovative robotic system designed for rehabilitation purposes. One of our key advancements is a treadmill equipped with additional degrees of freedom, enabling adjustments to its inclination in space. This feature allows targeted rehabilitation by requiring users to exert more effort on a specific leg, as guided by therapists.

My contributions include the development of C modules for the robot control software. These modules are designed to dynamically adjust the treadmill’s tilt and inclination to create controlled disequilibrium, synchronized with the virtual scene and therapist-defined corrections.

Additionally, I worked on a C# module responsible for communication between the scene generation software and the robot’s real-time parameters, such as speed, inclination, and skewness. This integration facilitates the generation of virtual terrain scenes, such as forest environments, aimed at engaging users and encouraging natural walking patterns during rehabilitation exercises.

Rehabilitation in an Immersive Environment

Rehabilitation can be a challenging and discomforting process, and tracking its progress effectively is often complicated. Incorporating immersive environments has demonstrated an increase in patient motivation, yet it does not fully address the need for enhancing rehabilitation efficiency. Visual feedback, even in 3D, alone is insufficient to create a fully immersive experience, as it lacks integration with physical body motion.

Body motion control is crucial for therapists, who are currently required to make constant adjustments to the patient’s posture to maximize the effectiveness of rehabilitation exercises. To address these limitations, we propose integrating motion generators into the immersive environment. This approach enhances realism—boosting patient motivation—and equips therapists with tools to guide body positioning, enabling repetitive rehabilitation exercises within a controlled framework.

In addition to guiding motion, these generators are equipped with instrumentation that gathers data on body posture. External sensors complement this setup, providing comprehensive measurements to evaluate rehabilitation progress. Our work includes the development of three types of motion generators: a six-degree-of-freedom motion base, a cable-driven parallel robot (CDPR) capable of lifting patients, and two versatile lifting columns.

Currently, we are merging visual feedback with these motion generators. However, progress has been delayed due to the unavailability of the immersive room at Inria-Sophia, which was a key element of our initial plan. This challenge necessitated the development of a custom renderer, slowing integration. Despite these setbacks, we have initiated preliminary experiments.

One experiment involves using the lifting columns to adjust the slope and inclination of a treadmill, simulating uphill walking scenarios. This setup allows for targeted rehabilitation, such as enforcing the use of a specific leg. External and wearable sensors monitor the walking pattern, providing synthesized data for therapists to assess and refine the rehabilitation process.