Sampling-based control improves safe robot tracking under uncertainty
CS-MPC: Contraction-Certified Sampling-Based MPC for Nonlinear Systems Under Uncertainty
Robotics
Summary
Robots that move around in the real world have to deal with unexpected problems or changes in their surroundings. This paper introduces a new way to help robots keep on track and avoid danger by planning many possible moves but focusing more on those that are safe and likely to work. The authors tested their method in computer simulations with rovers, drones, and landing robots, and it used fewer trial moves and stayed safer than older methods. The research shows promise for making robot control more reliable when conditions are uncertain.
What this means in practice
- •For robotics engineers: Improve real-time control of robots like rovers or drones to track targets safely with fewer calculations under uncertain conditions.
- •For space mission planners: Enhance safety and tracking accuracy for spacecraft landing or navigating in uncertain environments using improved control algorithms.
Authors
Yashwanth Kumar Nakka
Abstract
We present contraction-certified sampling-based model predictive control (CS-MPC) for tracking and safety of nonlinear robotic systems under parametric uncertainty, disturbances, and safety constraints that are revealed online. CS-MPC uses a contraction-certified tracking feedback inside every sampled rollout, which concentrates the rollouts around the reference, and an execution layer that verifies the applied input using a quantile-tightened discrete-time control barrier function condition. We derive a verification procedure for nonlinear contraction of the feedback-closed dynamics, and we prove a finite-sample stochastic stability bound and a mission-level probabilistic safety guarantee. We validate CS-MPC in simulation on a rover, a quadrotor, and an asteroid landing. When compared with standard model predictive path integral (MPPI) control, CS-MPC attains the same tracking error with 7 to 20 times fewer rollouts for the rover, and a 2 to 4 times lower tracking error for the quadrotor, with no safety violations in 200 Monte Carlo missions.