Stanford University
DexForce leverages contact forces, measured during kinesthetic demonstrations, to compute force-informed actions.
We use DexForce demonstrations to learn policies for6 tasksand evaluate each rollout according to the following criteria:
Policies trained on our force-informed actions achieve an average76%success rate across all 6 tasks.
Baseline policies trained on actions thatdo notaccount for contact forces havenear zerosuccess rates.
To view policy rollouts trainedwithandwithoutforce-informed actions for each task, view on larger screen.
Explore policies trainedwithandwithoutforce-informed actions for each task: