Multiple Behaviour In Autonomous Robotic Vehicle

摘要

The advancement in mobile robotics in recent decades have inspired and cemented a belief that multiple autonomous robotic agents, often cooperatively helping the humans. In order to provide more generalized adaptive capability in dynamic environments, it is desirable to exclude as many detail assumptions as possible. In order to utilize autonomous mobile robots in real life, the ability of adapting themselves to the environment which change in determinately. Studies have been conducted applying diverse algorithms to robot control for the learning of motion rules and path planning in general environments. . Utilizing DNA programming, Kozza showed that robots can find out motion roaming around grid spaces to get preys, with environmental information only. The robot's sensor information and position data relative to the obstacle is entered, classified to generate representative pattern, which in turn is entered into the Memory to generate generalized motion rules. The desired destination can be determined with information the robot generates.