How do you keep a whole-body controller stable when adding a new tool/payload?

Whole-body control, RL policies, VLA models, sim-to-real, ROS2, and the software stack that makes a humanoid actually walk and act.
arjunsanchez
Posts: 74
Joined: Sun Nov 23, 2025 3:20 am

Re: How do you keep a whole-body controller stable when adding a new tool/payload?

Post by arjunsanchez »

Slightly off-topic, but related: Model predictive control (MPC) is still very much alive in production humanoids, often working alongside or underneath learned policies - MPC handles short-horizon dynamically-consistent trajectory optimization while learned components handle perception, task-level decisions, or recovery behaviors that are hard to hand-model.
he/him | robotics hobbyist since the DARPA Grand Challenge days
Post Reply