Bi-Level Optimization for Contact and Motion Planning in Rope-Assisted Legged Robots

TL;DR

This paper introduces a bi-level optimization framework for planning contact and motion in rope-assisted legged robots climbing vertical surfaces, combining mixed-integer and gradient-based methods.

Contribution

It presents a novel planning pipeline that integrates bi-level optimization for terrain selection and control input optimization in rope-assisted climbing robots.

Findings

Validated on the ALPINE robot across various terrains.

Effectively combines mixed-integer and gradient-based optimization.

Demonstrates feasible and efficient climbing motions.

Abstract

This paper presents a planning pipeline framework for locomotion in rope-assisted robots climbing vertical surfaces. The proposed framework is formulated as a bi-level optimization scheme that addresses a mixed-integer problem: selecting feasible terrain regions for landing while simultaneously optimizing the control inputs, namely rope tensions and leg forces, and landing location. The outer level of the optimization is solved using the Cross-Entropy Method, while the inner level relies on gradient-based nonlinear optimization to compute dynamically feasible motions. The approach is validated on a novel climbing robot platform, ALPINE, across a variety of challenging terrain configurations.