CLIS-MPPI: Closed-Loop Model Predictive Path Integral Control With Initial-State Sampling Under Execution Delay
CLIS-MPPI addresses planning inaccuracies caused by execution delays in motion control. It models planning delay as uncertainty over initial states, samples these states for rollouts, and applies a feedback correction during execution. This closed-loop approach significantly improves safety and consistency compared to standard MPPI, resulting in fewer collisions and smoother paths.