loading page

The Verification and Testing on Reactive Robot Cooperating Systems
  • Saruipamade Pavakoamiea
Saruipamade Pavakoamiea
Namibe Research Institute of Engineering and Technology

Corresponding Author:[email protected]

Author Profile

Abstract

In this paper, an robot parallel evolution design algorithm is proposed, based on the idea of module network, and the learning method is verified and tested as well. The learning method of collision avoidance, approach and wall switching behavior of evolutionary robots based on neural network method is optimized. The evolutionary robot can autonomously realize the behaviors of collision avoidance, movement, replication and attack. In the simulation environment of this paper, the obstacles in the environment are represented by a randomly generated rectangular region with the largest side length. The shape and location of obstacles are evenly distributed. The simulated environment has an approach target but no obstacles, and these approach targets are placed randomly and scattered in the environment. The formation of behavior is integrated with basic behaviors such as collision avoidance and approach, so that the robot is gradually enhanced based on the growth of neural network.