Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Task-Level Decisions to Gait Level Control: A Hierarchical Policy Approach for Quadruped Navigation

Sijia Li, Haoyu Wang, Shenghai Yuan, Yizhuo Yang, Thien-Minh Nguyen

TL;DR

This paper presents a hierarchical policy architecture for quadrupedal navigation, termed Task-level Decision to Gait Control (TDGC), and introduces a structured curriculum with performance-driven progression that expands environmental difficulty and disturbance ranges.

Abstract

Real-world quadruped navigation is constrained by a scale mismatch between high-level navigation decisions and low-level gait execution, as well as by instabilities under out-of-distribution environmental changes. Such variations challenge sim-to-real transfer and can trigger falls when policies lack explicit interfaces for adaptation. In this paper, we present a hierarchical policy architecture for quadrupedal navigation, termed Task-level Decision to Gait Control (TDGC). A low-level policy, trained with reinforcement learning in simulation, delivers gait-conditioned locomotion and maps task requirements to a compact set of controllable behavior parameters, enabling robust mode generation and smooth switching. A high-level policy makes task-centric decisions from sparse semantic or geometric terrain cues and translates them into low-level targets, forming a traceable decision pipeline without dense maps or high-resolution terrain reconstruction. Different from end-to-end approaches, our architecture provides explicit interfaces for deployment-time tuning, fault diagnosis, and policy refinement. We introduce a structured curriculum with performance-driven progression that expands environmental difficulty and disturbance ranges. Experiments show higher task success rates on mixed terrains and out-of-distribution tests.

Task-Level Decisions to Gait Level Control: A Hierarchical Policy Approach for Quadruped Navigation

TL;DR

This paper presents a hierarchical policy architecture for quadrupedal navigation, termed Task-level Decision to Gait Control (TDGC), and introduces a structured curriculum with performance-driven progression that expands environmental difficulty and disturbance ranges.

Abstract

Real-world quadruped navigation is constrained by a scale mismatch between high-level navigation decisions and low-level gait execution, as well as by instabilities under out-of-distribution environmental changes. Such variations challenge sim-to-real transfer and can trigger falls when policies lack explicit interfaces for adaptation. In this paper, we present a hierarchical policy architecture for quadrupedal navigation, termed Task-level Decision to Gait Control (TDGC). A low-level policy, trained with reinforcement learning in simulation, delivers gait-conditioned locomotion and maps task requirements to a compact set of controllable behavior parameters, enabling robust mode generation and smooth switching. A high-level policy makes task-centric decisions from sparse semantic or geometric terrain cues and translates them into low-level targets, forming a traceable decision pipeline without dense maps or high-resolution terrain reconstruction. Different from end-to-end approaches, our architecture provides explicit interfaces for deployment-time tuning, fault diagnosis, and policy refinement. We introduce a structured curriculum with performance-driven progression that expands environmental difficulty and disturbance ranges. Experiments show higher task success rates on mixed terrains and out-of-distribution tests.
Paper Structure (19 sections, 34 equations, 2 figures, 1 table, 1 algorithm)

This paper contains 19 sections, 34 equations, 2 figures, 1 table, 1 algorithm.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORKS
  3. Learning Based Locomotion Control for Quadrupeds
  4. Task-Level Navigation for Legged Robots
  5. Hierarchical Policies and Training Strategies
  6. PRELIMINARIES
  7. Task Setup and Terrain Curriculum
  8. MDP Formulation
  9. METHODOLOGY
  10. Low-Level Gait-Conditioned Controller
  11. High-Level Task Policy
  12. Observation and action
  13. Command decoding and gait discretization
  14. Reward design
  15. Curriculum Mechanism
  16. ...and 4 more sections

Figures (2)

  • Figure 1: Hierarchical navigation framework. (A) The high-level recurrent policy maps task/terrain cues to a compact behavior output, which is decoded into an executable command and a discrete gait selection (trot, pronk, pace, bound). (B) The low-level gait-conditioned policy uses proprioception, the decoded command, and short action history to generate joint-level actions for stable locomotion. (C) The two policies run in closed loop with proprioceptive feedback to support goal-reaching across diverse terrains.
  • Figure 2: Qualitative rollouts on hard terrain levels (6--10) across five terrain families, comparing GP and TDGC.