01.08.2026
The Intelligence Layer
A powered joint gives us the ability to supply energy. Making that energy useful requires sensing and control that can respond to the person and the movement happening around it. We’re developing those capabilities together as part of the Genesis system.
The hardware combines multi-axis inertial sensing, actuator feedback, and onboard computing. Those signals give the controller information about how the limb is moving and being loaded. The controller uses that information to adjust output through the gait cycle, with the aim of making the joint’s contribution fit the movement.
ADA, our Adaptive Drive Architecture, sits at the center of that work. It processes live motion signals and determines how the actuator should respond from moment to moment. We’re using it to develop the timing and control of assistance through push-off, stance, and changes in terrain. Reducing how much attention a person needs to devote to managing the limb is one goal of this work.
Developing that behavior requires data we can use consistently. Our units in controlled field deployment collect high-frequency motion data during use. GDS, Genesis Data Systems, is our internal framework for modeling and validation. It supports analysis of individual gait patterns, terrain transitions, and difficult operating conditions, as well as regression testing and model refinement.
The purpose of collecting those data is to examine how a controller behaves across different people, speeds, surfaces, and loads. A proposed change needs to be evaluated against that evidence. We need to understand where it helps, where it does not, and whether it introduces a problem elsewhere in the system.
Core movement control stays onboard. Connectivity supports diagnostics and development, while the device retains local responsibility for the control needed during ambulation. That separation is important for a system that has to keep responding as the person moves.
The same architecture lets us refine control software after the hardware has been built. Those refinements need internal validation and a structured update process. Over time, we want the sensing, modeling, and control work to support additional Genesis configurations, so that each hardware iteration can build on what we have already learned.
The intelligence layer is how we connect the joint’s physical capabilities to the person using them. Its value will be in how well that connection works in practice.