Systematic study of the performance of small robots on controlled laboratory substrates
Systematic study of the performance of small robots on controlled laboratory substrates
复制标题
小型机器人在受控实验室基材上性能的系统研究
DOI：
10.1117/12.851047
复制
发表时间：
2010
期刊：
影响因子：
25
通讯作者：
D. Goldman
中科院分区：
文献类型：
--
作者：
Chen Li;Aaron M. Hoover;Paul Birkmeyer;P. Umbanhowar;R. Fearing;D. Goldman
设计的机器人能够有效地跨越各种地形需要详细的地面相互作用模型，不幸的是，这样的模型是缺乏的，由于复杂的反应，真实的世界基板，可以产生和流动的负载。为了促进我们对相关建模和设计问题的理解，我们对DASH和RoACH的性能进行了比较研究，这两个小型的，生物启发的，六条腿的，轻量级（~10 cm，~20 g）的机器人使用智能复合材料微结构（SCM）工艺制造。我们系统地研究这两个机器人在刚性和流动基板上的性能。改变地面属性和肢体步频，我们调查平均速度，平均机械功率和运输成本，和稳定性。我们发现，机器人的性能和稳定性是敏感的物理地面相互作用：在硬地面上的动能必须管理，以防止偏航，俯仰和滚动不稳定，以保持高性能，而在沙子上的流化相互作用导致运输成本增加和运行速度降低。我们还观察到，每个机器人的特征肢体形态和运动学的结果在他们的能力，穿越不同的地形有明显的差异。我们的系统研究是第一步，发展模型的相互作用的肢体与复杂的地形，以及开发改进的肢体形态和控制策略。
The design of robots able to locomote effectively over a diversity of terrain requires detailed ground interaction models; unfortunately such models are lacking due to the complicated response of real world substrates which can yield and flow in response to loading. To advance our understanding of the relevant modeling and design issues, we conduct a comparative study of the performance of DASH and RoACH, two small, biologically inspired, six legged, lightweight (~10 cm, ~20 g) robots fabricated using the smart composite microstructure (SCM) process. We systematically examine performance of both robots on rigid and flowing substrates. Varying both ground properties and limb stride frequency, we investigate average speed, mean mechanical power and cost of transport, and stability. We find that robot performance and stability is sensitive to the physics of ground interaction: on hard ground kinetic energy must be managed to prevent yaw, pitch, and roll instability to maintain high performance, while on sand the fluidizing interaction leads to increased cost of transport and lower running speeds. We also observe that the characteristic limb morphology and kinematics of each robot result in distinct differences in their abilities to traverse different terrains. Our systematic studies are the first step toward developing models of interaction of limbs with complex terrain as well as developing improved limb morphologies and control strategies.