Reactive Motion Generation With Particle-Based Perception in Dynamic Environments
🤖 AI Summary
This work proposes an obstacle-aware Model Predictive Path Integral (MPPI) planning framework to address the challenges posed by dynamic, unstructured environments, where traditional methods struggle due to their reliance on static perception and inability to reliably model dynamic obstacles under perception and control uncertainties. The approach explicitly maintains estimates of dynamic obstacles’ velocities and their covariances through a tensorized particle weight update mechanism and jointly propagates the dynamics of both the robot and obstacles to anticipate future interactions. By tightly integrating particle-based dynamic mapping with reactive planning, the framework significantly enhances safety and responsiveness in uncertain settings. Extensive simulations and real-world experiments with sensor noise demonstrate that the proposed method substantially outperforms existing MPPI-based baselines in navigation tasks involving multiple static and dynamic obstacles.
Application Category
📝 Abstract
Reactive motion generation in dynamic and unstructured scenarios is typically subject to essentially static perception and system dynamics. Reliably modeling dynamic obstacles and optimizing collision-free trajectories under perceptive and control uncertainty are challenging. This article focuses on revealing tight connection between reactive planning and dynamic mapping for manipulators from a model-based perspective. To enable efficient particle-based perception with expressively dynamic property, we present a tensorized particle weight update scheme that explicitly maintains obstacle velocities and covariance meanwhile. Building upon this dynamic representation, we propose an obstacle-aware MPPI-based planning formulation that jointly propagates robot-obstacle dynamics, allowing future system motion to be predicted and evaluated under uncertainty. The model predictive method is shown to significantly improve safety and reactivity with dynamic surroundings. By applying our complete framework in simulated and noisy real-world environments, we demonstrate that explicit modeling of robot-obstacle dynamics consistently enhances performance over state-of-the-art MPPI-based perception-planning baselines avoiding multiple static and dynamic obstacles.
Problem
Research questions and friction points this paper is trying to address.
reactive motion generation
dynamic environments
particle-based perception
collision-free trajectories
perception uncertainty
Innovation
Methods, ideas, or system contributions that make the work stand out.
particle-based perception
dynamic obstacle modeling
MPPI-based planning
tensorized weight update
reactive motion generation
🔎 Similar Papers
No similar papers found.
