Affiliation:
1. School of Aerospace Engineering and Applied Mechanics, Tongji University, China
2. Department of Mechanical Engineering, University of Michigan, USA
3. Institute of AI and Robotics, Fudan University, China
Abstract
The goal of this research is to develop a generic earthworm-like locomotion robot model consisting of a large number of segments in series and based on which to systematically investigate the generation of planar locomotion gaits and their correlation with a robot’s locomotion performance. The investigation advances the state-of-the-art by addressing some fundamental but largely unaddressed issues in the field. These issues include (a) how to extract the main shape and deformation characteristics of the earthworm’s body and build a generic model, (b) how to coordinate the deformations of different segments such that steady-state planar locomotion can be achieved, and (c) how different locomotion gaits would qualitatively and quantitatively affect the robot’s locomotion performance, and how to evaluate them. Learning from earthworms’ unique morphology characteristics, a generic kinematic model of earthworm-like metameric locomotion robots is developed. Left/right-contracted segments are introduced into the model to achieve planar locomotion. Then, this paper proposes a gait-generation algorithm by mimicking the earthworm’s retrograde peristalsis wave, with which all admissible locomotion gaits can be constructed. We discover that when controlled by different gaits, the robot would exhibit four qualitatively different locomotion modes, namely, rectilinear, sidewinding, circular, and cycloid locomotion. For each mode, kinematic indexes are defined and examined to characterize their locomotion performances. For verification, a proof-of-concept robot hardware is designed and prototyped. Experiments reveal that with the proposed robot model and the employed gait controls, locomotion of different modes can be effectively achieved, and they agree well with the theoretical predictions.
Funder
national natural science foundation of china
University of Michigan Collegiate Professorship
Subject
Applied Mathematics,Artificial Intelligence,Electrical and Electronic Engineering,Mechanical Engineering,Modelling and Simulation,Software
Cited by
32 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献