At the European Robotics Forum in Stavanger, we caught up with Jean-François de Sallier to see how force-feedback wearables and optical sensing are converging to accelerate robot learning. This year, the French haptics specialist has taken its signature force-feedback technology into new territory — pairing it with SenseGlove’s wearable optical sensors to create a combined data-capture and teleoperation platform that could reshape how humanoid robots learn.
Jean-François de Sallier walked us through the setup. The new system layers Haption’s per-finger force feedback — which lets an operator physically feel the contact forces at the very tip of each finger — with high-speed wearable optics that capture hand and finger movement data at up to one millisecond resolution. That combination, he explained, is what makes the platform stand out.
The big problem: teaching robots quickly
The timing is deliberate. With humanoid robots dominating headlines — and conference hallways — one of the field’s pressing challenges is imitation learning: getting a robot to observe a task performed by a human and replicate it reliably. De Sallier sees the Haption–SenseGlove platform as a direct answer to that challenge.
“Today there are lots of topics about humanoids, about robotics,” he said, “and one of the big problems is how to learn quickly. You can really showcase immediately to the robots how to take an object, how to solve an issue — and then, with data collection in real time, you have exactly that setup of acquisition.”
The idea is that an operator wearing the gloves performs a task in a natural, intuitive way — guided by haptic feedback to replicate real interaction forces — while SenseGlove’s optical sensors capture every nuance of the movement. That rich, timestamped dataset can then be fed directly into robot learning pipelines.
