HiPAN: Hierarchical Posture-Adaptive Navigation for Quadruped Robots in Unstructured 3D Environments

arXiv cs.RO / 4/30/2026

📰 NewsDeveloper Stack & InfrastructureIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes HiPAN, a hierarchical navigation framework for quadruped robots operating in unstructured 3D spaces with narrow, height-constrained passages.
Instead of a multi-stage mapping–planning pipeline, HiPAN uses onboard depth images at deployment and splits the task into a high-level policy (planar velocity and body posture) plus a low-level posture-adaptive locomotion controller.
To improve long-horizon navigation and avoid myopic local behaviors, it introduces Path-Guided Curriculum Learning that gradually expands the navigation horizon from reactive avoidance to strategic goal-directed movement.
Simulation results show higher success rates and better path efficiency than classical reactive planners and end-to-end baselines, and real-world experiments support its effectiveness across varied environments.
The approach targets reduced computational overhead and mitigates accumulated perception errors, aiming to be more usable on resource-constrained platforms.

Abstract

Navigating quadruped robots in unstructured 3D environments poses significant challenges, requiring goal-directed motion, effective exploration to escape from local minima, and posture adaptation to traverse narrow, height-constrained spaces. Conventional approaches employ a sequential mapping-planning pipeline but suffer from accumulated perception errors and high computational overhead, restricting their applicability on resource-constrained platforms. To address these challenges, we propose Hierarchical Posture-Adaptive Navigation (HiPAN), a framework that operates directly on onboard depth images at deployment. HiPAN adopts a hierarchical design: a high-level policy generates strategic navigation commands (planar velocity and body posture), which are executed by a low-level, posture-adaptive locomotion controller. To mitigate myopic behaviors and facilitate long-horizon navigation, we introduce Path-Guided Curriculum Learning, which progressively extends the navigation horizon from reactive obstacle avoidance to strategic navigation. In simulation, HiPAN achieves higher navigation success rates and greater path efficiency than classical reactive planners and end-to-end baselines, while real-world experiments further validate its applicability across diverse, unstructured 3D environments.