Neuromodulatory developmental learning of the mobile robots corresponding to the unexpected obstacles

With the gradual expansion of robot applications, the operating environment is becoming more and more complex, and various uncertainty may be encountered. Investigating how to efficiently respond to various uncertainty in the environment has become an important challenge in the field of robotics research. For the autonomous obstacle avoidance of mobile robots in case of sudden appeared obstacles, a dynamic obstacle avoidance algorithm with a motivated developmental network that simulates the visual attention mechanism is proposed. Simulating the response mechanism of biological vision, a depth camera is used to achieve the detection and recognition of obstacles. To enhance the behavioral regulation of mobile robots, the response mechanism of the human brain attention network is simulated, and an attention model containing the ventral attention network and dorsal attention network is proposed, then a motivated developmental network is designed to simulate this attention mechanism. Furthermore, the working mechanism of the neuromodulation system is simulated to better regulate the robot’s motion and improve its ability to quickly respond to dynamic obstacles suddenly appeared in the environment. A new collision risk is designed by considering the influence of the obstacle’s speed, direction, and distance to the mobile robot. Finally, the feasibility of the proposed method is verified by the experimental results in different physical environments.