Design of Positioning Accuracy Improvement System for Robot Navigation Based on Dead Reckoning Collaboration

Intelligent robots can be widely used in various scenes, and autonomous navigation is one of its key technical elements. Collision avoidance warning in autonomous movement of robot clusters has been widely concerned in various applications. The existing solution is to constantly broadcast its position to nearby robots through robot wireless communication technology. In the traditional sense, the positioning accuracy of robots largely depends on GPS equipment. However, there are still some factors affecting the positioning accuracy of GPS equipment. Conventional GPS can provide accuracy within 5–30 meters in a dense urban environment, which is not enough to support safe navigating of robots. This paper designs a positioning accuracy improvement system for the robot swarm based on dead reckoning collaboration. The DR algorithm is used to correct the GPS position of the target robot and adjacent robots, and the IPC algorithm is used to correct the initial position deviation in the DR algorithm. The combination of IPC algorithm and DR algorithm can effectively reduce the GPS positioning error, and does not need additional information such as city digital maps. In the actual urban environment, the average positioning accuracy of this method can be improved by about 15%.