CSR:Small: TEAMS -- Transplanting Artificial Life Behaviors to Mobile Robots

点击翻译按钮获取中文摘要

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).The TEAMS project applies digital evolution to the design of robust communication services for cooperating groups of robots. Example applications include teams of robots for disaster relief operations, assisting humans in dangerous occupations, and monitoring of critical infrastructure and public service facilities such as drinking water reservoirs. In digital evolution, a population of self-replicating computer programs exists in a user-defined computational environment and is subject to mutations and natural selection. Evolved algorithms (sequences of instructions comprising the genomes of digital organisms) can be recompiled to execute directly on hardware platforms. The project explores three overlapping classes of behavior: distributed self-organization of nodes; collective communication operations that tolerate adverse conditions; and mobility-aided communication methods, where nodes change their locations in order to improve communication performance. Evolved behaviors are evaluated on an NSF-sponsored digital evolution testbed containing a heterogeneous collection of microrobots. The ability to observe the evolutionary process in digital organisms provides a powerful method to investigate the driving forces in complex systems. The TEAMS project exploits this open-ended method of exploration to find better ways to construct software for mobile robotic systems that need to interact with the physical world, and each other, in order to solve problems. On a broader scale, the project includes several innovative educational and outreach components, such as an after-school program to provide economically disadvantaged K-12 students with exciting hands-on experiences involving evolutionary computation, adaptive software, and robotics.