Event
Proposes ContactFlowFilter online contact estimation method
1 reports · 1 sources
Seo Wook Han and Min Jun Kim propose ContactFlowFilter for estimating the number, locations, and force vectors of active contacts on a robotic arm equipped only with joint torque sensors and a base force/torque sensor. The method uses a flow matching model as the proposal distribution of a particle filter, employs the recursive structure of the particle filter to capture temporal dependencies, tracks multimodal beliefs, and detects contact addition and removal events. The paper states that the method was validated in simulation and real-world experiments; in real-world scenarios, the position errors for single contact and dual contact were 0.36 cm and 0.88 cm, respectively, with runtimes of 0.69 ms and 2.69 ms.
Report timeline
- ContactFlowFilter: An Online Contact Estimation Method for Robotic Arms Based on Flow Matching and Particle Filtering
The paper proposes ContactFlowFilter for estimating the number, locations, and force vectors of active contacts on a robotic arm equipped only with joint torque sensors and a base force/torque sensor. The method uses a flow matching model as an auxiliary proposal distribution for a particle filter, leveraging the particle filter's recursive structure to capture temporal dependencies while tracking multimodal beliefs and detecting contact addition and removal events.
arXiv cs.RO ·