An Evaluation Methodology for Autonomous Mobile Robots for Planetary Exploration
Gaurav S. Sukhatme, George A. Bekey
- 发表年份
- 1995
- 引用次数
- 10
摘要
We report here on a new technique to evaluate autonomous mobile robots built for exploring rugged terrain such as is found on planetary surfaces. The dynamics of both wheeled and legged robots interacting with deformable soil (sand) and obstacles are notoriously complex and nonlinear. This paper introduces a novel statistical methodology for the evaluation of such robots. We report on work in progress using a wheeled and a legged robot as examples. We have performed experiments in simulation. Predictions from these are reported. We have built two robots and a testbed which simulates Martian terrain for testing these predictions. 1 Introduction Robots are complex systems, whose interaction with the environment is hard to model accurately. The question of whether one has the appropriate robot for a given mission is often asked but seldom answered. Given two robot designs a comparison between the two is difficult because of the many (often incommensurate) dimensions along which performan...
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002