Publication Date


Technical Report Number



Computing Methodologies


Although evolutionary algorithms offer an attractive and versatile approach to the automated design of behavior and control structures for mobile robots, they cannot anticipate the detailed structure of specific environments that the robot might have to deal with. Robots must thus possess mechanisms to adapt to the environments they encounter. In particular, mobile robots need structures for building and using spatial maps to aid in the successful exploration and navigation of a-priori unknown environments. This paper proposes a biologically inspired computational model for the acquisition and use of spatial memory structures for mobile robot navigation. Preliminary experimental results indicate that the proposed mechanisms can be effectively exploited by evolution in the design of high-performance robots