Decremental Dynamics Planning for Robot Navigation

TL;DR

This work addresses the mismatch between global and local planning caused by assuming either full or no robot dynamics in hierarchical navigation. It proposes Decremental Dynamics Planning (DDP), which starts with high-fidelity dynamics in early trajectory rollout and gradually reduces fidelity to meet real-time constraints. The authors demonstrate DDP by augmenting three planning algorithms (DWA, MPPI, Log-MPPI) and by implementing a standalone DDP-based navigation system, achieving strong performance in simulated and physical experiments, including the 2025 BARN Challenge. The results show improved planning success rates, reduced collisions, and competitive computational efficiency, suggesting DDP as a general framework for robust, real-time robot navigation.

Abstract

Most, if not all, robot navigation systems employ a decomposed planning framework that includes global and local planning. To trade-off onboard computation and plan quality, current systems have to limit all robot dynamics considerations only within the local planner, while leveraging an extremely simplified robot representation (e.g., a point-mass holonomic model without dynamics) in the global level. However, such an artificial decomposition based on either full or zero consideration of robot dynamics can lead to gaps between the two levels, e.g., a global path based on a holonomic point-mass model may not be realizable by a non-holonomic robot, especially in highly constrained obstacle environments. Motivated by such a limitation, we propose a novel paradigm, Decremental Dynamics Planning that integrates dynamic constraints into the entire planning process, with a focus on high-fidelity dynamics modeling at the beginning and a gradual fidelity reduction as the planning progresses. To validate the effectiveness of this paradigm, we augment three different planners with DDP and show overall improved planning performance. We also develop a new DDP-based navigation system, which achieves first place in the simulation phase of the 2025 BARN Challenge. Both simulated and physical experiments validate DDP's hypothesized benefits.

Figures (5)

• Figure 1: Contrasting the traditional global and local planning paradigm (top), where either full (green) or zero (white) robot dynamics is considered, DDP starts with high fidelity dynamics in the early part of trajectory rollout and gradually decreases dynamics fidelity for computation efficiency (bottom).
• Figure 2: An example of our standalone DDP-based system navigating a BARN environment. Left: Visualization in RViz. Right: Visualization in Gazebo.
• Figure 3: Standalone DDP-based Navigation System.
• Figure 4: Two Physical Test Environments.
• Figure 5: Experiments in Real-World, Natural Cluttered Spaces.