[Paper Review] Underactuated Motion Planning and Control for Jumping with Wheeled-Bipedal Robots

Hua Chen, Bingheng Wang+4 more|arXiv (Cornell University)|Dec 11, 2020

Robotic Locomotion and Control31 references4 citations

TL;DR

This paper proposes a hierarchical motion planning and control framework for jumping in wheeled-bipedal robots using a novel wheeled-spring-loaded inverted pendulum (W-SLIP) model. Leveraging a differential-flatness-like property, it formulates quadratic programs for stance and flight phase planning, and integrates a disturbance observer-based composite controller for accurate trajectory tracking, enabling successful simulation-based jumping across diverse conditions including varied heights and sloped landings.

ABSTRACT

This paper studies jumping for wheeled-bipedal robots, a motion that takes full advantage of the benefits from the hybrid wheeled and legged design features. A comprehensive hierarchical scheme for motion planning and control of jumping with wheeled-bipedal robots is developed. Underactuation of the wheeled-bipedal dynamics is the main difficulty to be addressed, especially in the planning problem. To tackle this issue, a novel wheeled-spring-loaded inverted pendulum (W-SLIP) model is proposed to characterize the essential dynamics of wheeled-bipedal robots during jumping. Relying on a differential-flatness-like property of the W-SLIP model, a tractable quadratic programming based solution is devised for planning jumping motions for wheeled-bipedal robots. Combined with a kinematic planning scheme accounting for the flight phase motion, a complete planning scheme for the W-SLIP model is developed. To enable accurate tracking of the planned trajectories, a linear quadratic regulator based wheel controller and a task-space whole-body controller for the other joints are blended through disturbance observers. The overall planning and control scheme is validated using V-REP simulations of a prototype wheeled-bipedal robot.

Motivation & Objective

• To address the lack of comprehensive planning and control strategies for agile motions in underactuated wheeled-bipedal robots.
• To develop a unified motion planning framework that handles the transition between flight and stance phases during jumping.
• To overcome the challenges posed by underactuated dynamics in the stance phase of wheeled-bipedal robots.
• To enable high-precision trajectory tracking despite model-actuator discrepancies and disturbances.
• To validate the framework on a simulated wheeled-bipedal robot across diverse jumping scenarios, including sloped landings.

Proposed method

• Introduces the wheeled-spring-loaded inverted pendulum (W-SLIP) model to capture essential dynamics of wheeled-bipedal robots during jumping.
• Exploits a differential-flatness-like property of the W-SLIP model to enable tractable quadratic programming (QP)-based motion planning for the stance phase.
• Uses a kinematic model of the W-SLIP for flight phase planning via another QP formulation to ensure smooth transitions.
• Designs a disturbance observer-based composite controller combining a linear quadratic regulator (LQR) for wheel control and a task-space whole-body controller for other joints.
• Implements switching logic based on wheel velocity to trigger stance phase planning during landing.
• Validates the entire framework using V-REP simulations on the SUSTech Nezha robot with a 10 Hz sampling rate.

Experimental results

Research questions

• RQ1 How can the underactuated dynamics of wheeled-bipedal robots be effectively modeled for jumping motion planning?
• RQ2 Can a differential-flatness-like property in the W-SLIP model enable efficient and tractable trajectory planning via quadratic programming?
• RQ3 How can flight phase motion be planned in coordination with stance phase planning to ensure stable takeoff and landing?
• RQ4 Can a composite controller with disturbance observers achieve accurate tracking of planned trajectories despite model inaccuracies?
• RQ5 What is the robustness and performance envelope of the proposed framework across varying initial velocities and desired jump heights?

Key findings

• The proposed W-SLIP model successfully captures the essential dynamics of wheeled-bipedal robots during jumping, enabling effective planning.
• The QP-based planning scheme covers a wide region of attraction in the state space, particularly failing only when tilting angle, tilting rate, and leg length rate are all high.
• Simulations on the SUSTech Nezha robot demonstrate successful jumps with desired center-of-mass apex heights ranging from 0.8 m to 1.4 m.
• The framework enables successful jumps with initial forward velocities of 0.5 m/s, 1.0 m/s, and 1.5 m/s, confirming robustness to velocity variations.
• A successful jump on a 5° inclined slope was achieved, demonstrating the framework’s robustness to uneven terrain.
• The actual apex height achieved was slightly lower than the desired value (e.g., 1.4 m desired, lower actual), attributed to discrepancies between the W-SLIP model and the full multi-link robot model.