Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Reduced-Order Model-Based Gait Generation for Snake Robot Locomotion using NMPC

Adarsh Salagame, Eric Sihite, Milad Ramezani, Alireza Ramezani

TL;DR

This work addresses gait generation for snake robots in constrained environments by combining a reduced-order model with nonlinear model predictive control (NMPC). The ROM abstracts the robot to a center-of-mass frame with a bounding-box footprint and a contact signal, enabling efficient path planning within a safety corridor inflated along a nominal path. NMPC optimizes the central pattern generator (CPG) parameters and the center-of-mass trajectory to minimize tracking error while respecting footprint constraints and contact dynamics, using high-fidelity simulations and hardware experiments to validate the approach. The results demonstrate autonomous gait generation that preserves footprint safety and enables navigation through narrow corridors, contributing to practical autonomy for snake-robot locomotion in confined spaces.

Abstract

This paper presents an optimization-based motion planning methodology for snake robots operating in constrained environments. By using a reduced-order model, the proposed approach simplifies the planning process, enabling the optimizer to autonomously generate gaits while constraining the robot's footprint within tight spaces. The method is validated through high-fidelity simulations that accurately model contact dynamics and the robot's motion. Key locomotion strategies are identified and further demonstrated through hardware experiments, including successful navigation through narrow corridors.

Reduced-Order Model-Based Gait Generation for Snake Robot Locomotion using NMPC

TL;DR

This work addresses gait generation for snake robots in constrained environments by combining a reduced-order model with nonlinear model predictive control (NMPC). The ROM abstracts the robot to a center-of-mass frame with a bounding-box footprint and a contact signal, enabling efficient path planning within a safety corridor inflated along a nominal path. NMPC optimizes the central pattern generator (CPG) parameters and the center-of-mass trajectory to minimize tracking error while respecting footprint constraints and contact dynamics, using high-fidelity simulations and hardware experiments to validate the approach. The results demonstrate autonomous gait generation that preserves footprint safety and enables navigation through narrow corridors, contributing to practical autonomy for snake-robot locomotion in confined spaces.

Abstract

This paper presents an optimization-based motion planning methodology for snake robots operating in constrained environments. By using a reduced-order model, the proposed approach simplifies the planning process, enabling the optimizer to autonomously generate gaits while constraining the robot's footprint within tight spaces. The method is validated through high-fidelity simulations that accurately model contact dynamics and the robot's motion. Key locomotion strategies are identified and further demonstrated through hardware experiments, including successful navigation through narrow corridors.

Paper Structure

This paper contains 11 sections, 16 equations, 11 figures.

Table of Contents

  1. INTRODUCTION
  2. Problem Statement
  3. Reduced-Order Modeling
  4. Floating Base Frame for ROM
  5. Bounding Box Definition of ROM
  6. Contact Signal Estimation
  7. Low-level Joint Control
  8. Kinematic Motion Model
  9. Nonlinear Model Predictive Control (NMPC)
  10. Results
  11. Concluding Remarks

Figures (11)

  • Figure 1: Shows COBRA salagame_loco-manipulation_2024jiang_snake_2024jiang_hierarchical_2024salagame_heading_2024salagame_how_2024 traversing through a narrow path.
  • Figure 2: Highlights our approach to formulating COBRA's path planning and tracking problem by reducing its multi-DOF posture manipulation to a few key actions, including shape adjustment, forward and lateral translation, and rotation, all within a bounding box that encapsulates COBRA.
  • Figure 3: Shows the Reduced-Order Model for our snake robot COBRA
  • Figure 4: Depicts kinematic configuration and CPG parameters
  • Figure 5: Shows an overview of the system architecture
  • ...and 6 more figures