Target Position Estimation Aided Swarm Robotic Search under Conditions of Relative Localization Mechanism

Swarm robotic which search target with swarm intelligence are controlled in a coordinated way, the robot's own common knowledge and his group experience guide his behaviors. In essence, the group experience is the best one in the group of all robots' own common knowledge. To improve the speed of searching target, swarm robotic add the estimated the value of target's position under the relative localization mechanism, in order to give full play to the advantages of group decision. First, swarm robotic can take the model of extended particle swarm optimization (PSO) as a controlled tool. Then, based on the similarity between swarm robotic and wireless sensor networks and the nature of swarm robotic estimated the value of target's position with Received Signal Strength Indicator (RSSI), swarm robotic combine their experience and group decision-making to searching the target. When the robot can estimate the target position, the value of target's position will be introduced into the extended PSO model, Otherwise, it use the original model and they control the robots alternatingly. The experiment result proves that the new model is better than the old one among the success rates, the steps for searching target and energy consumption.