Whole-Body Motion Planning for Walking Excavators

Abstract

This article presents a trajectory planning framework for all-terrain vehicles with legs and wheels such as walking excavators. Our formulation takes into account the whole body of the robot while computing the plans for locomotion. Hence, we can produce motion plans over the rough terrain that would be hard to plan without considering all Degrees of Freedom (DoF) simultaneously. Our planner can also optimize over the contact schedule for all limbs, thereby finding the feasible motions even for the infeasible initial contact schedule. Furthermore, we introduce a novel formulation of the support area constraint. We generate plans for a Menzi Muck M545, a 31 DoF walking excavator with five limbs: four wheeled legs and an arm. We show motion plans for traversing a variety of terrains that require whole-body planning. To the best of our knowledge, this is the first work that addresses motion planning in rough terrain for vehicles with legs and wheels.

Terrain-Adaptive Planning and Control of Complex Motions for Walking Excavators

“…In [8], [9], ANYmal solves an Nonlinear Program (NLP) with box constraints on the end-effector position over a prediction horizon of 0.85 s -2 s. While motions on the real system look quite impressive, the planning update rates achieved (about 200 Hz) are chiefly thanks to the short prediction horizon and linear inequalities (box constraint) that can be used to enforce end-effector range of motion. In contrast, the kinematics of a walking excavator (absence of the knee joint) does not permit such a simplification, and one has to solve for the joint positions as well (see [10]). This renders the NLP considerably harder, which then results in a decreased Real-time (RT) factor.…”

“…In particular, we focus on large scale robots with many DoFs such as walking excavators. Our contributions can be summarized as follows: We extend the motion planner introduced in our previous work [10] for execution on the real hardware. Besides, we design a tracking controller for executing challenging whole-body motions on a real machine.…”

“…II. PLANNING In our previous work [10], we introduced a collocation [21] based planner that solves an optimal control problem. It produces kinematically consistent plans while respecting the non-holonomic rolling constraint for the wheels.…”

Trajectory Optimization for Hybrid Wheeled-Legged Robots in Challenging Terrain

“…On the other hand, the motion planner presented by (Jelavic and Hutter, 2019) solves the whole-body planning problem combining driving and stepping motions, also allowing for gait optimization. The framework, however, focuses on generating kinematically feasible motions for wheeled-legged vehicles performing slow maneuvers, which is the case for heavy machines such as walking excavators.…”

“…To overcome this problem, the abrupt changes in the terrain are smoothed to ensure convergence. For the optimization, the gap was modeled as a 1.5 m deep parabola, as exemplified in (Jelavic and Hutter, 2019). The same procedure is done for floating steps by inserting a linear transition of a very steep slope (∼ 10) between the ground and the step.…”