A benchmark for measuring mobile robot environment modelling performance
暂无分享,去创建一个
This paper presents a novel benchmarking toolkit suitable for quantitatively evaluating the performance of mobile robot environment modelling methods, thus allowing comparisons to be made. The benchmark measure is based on re-localisation, itself a practical navigation task and is demonstrated on a biomimetic survey navigation model developed within our research group and implemented on an autonomous mobile robot. The results show that the benchmark can usefully quantify an implementation of the model, based on predefined criteria, and can be used to find optimal model parameters.
[1] K. Cheng. A purely geometric module in the rat's spatial representation , 1986, Cognition.
[2] Michael Recce,et al. Towards intelligent mobile robots-scientific methods in mobile robotics , 1997 .
[3] E. Tolman. Cognitive maps in rats and men. , 1948, Psychological review.
[4] Vic Callaghan,et al. From Mammals to Machines: Towards A Biologically Inspired Mapping Model For Autonomous Mobile Robots , 2000 .