A navigation method for autonomous guided vehicles in industrial environments

• Main Author: Huang, He
• Other Authors: Xie Lihua
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/140897
• Institution: Nanyang Technological University

Navigation and obstacle avoidance of industrial robots are the key to the automation, and the realization of these functions depends on reliable sensors and corresponding algorithms. In the increasingly complex industrial environment, a single sensor can no longer meet the needs, so the use of multi-sensor fusion becomes a better choice, based on which the navigation and obstacle avoidance of robots can be realized. This dissertation presents a sensor fusion based control model for industrial robot navigation and obstacle avoidance. Firstly, we analyze the positioning methods, characteristics and error comparison of different sensors. Then the Extended Kalman Filter (EKF) is used to fuse UWB data and IMU data to overcome the disadvantages of single sensors. Simulation results show that this method could improve the accuracy of position estimation and reduce positioning errors. A controller is designed for the navigation task which consists of the linear velocity control, rotating velocity control and an amplitude limit to ensure safety. Then it is extended in the case of obstacle avoidance and the controller design is transformed into an optimization problem and how to seek the global optimal solution. For the implementation of the method, we design the software for this model, which contains the application of ROS framework, Finite Control Machine (FSM) and PID controller for the chassis. Finally, in order to verify the effectiveness of the model, we design an experiment under the simulation scene. As a consequence, the industrial robot moves from an arbitrary location to the target station and delivers items, which validates the correctness and feasibility of the model