Haopeng He, Wei Xiang, and Haibo Liu
Multi sensor, autonomous navigation, path planning, extended Kalman filtering, ant colony
For service robots, the most important thing is to solve how to identify the surrounding environment, how to move to a target location, and how to avoid obstacles. Therefore, the research first fuses the depth camera and the laser radar, and uses the extended Kalman filter (EKF) method to optimise it. The maximum as well as minimum amplitude values after fusion are 1.472 and 1.128, respectively. A comparison of the amplitude trend after integration with the true value shows that the integrated amplitude is closer to the actual value. Then, an improved ant colony algorithm (ACA) is studied to optimise path planning (PP). Compared with ACA, this algorithm outperforms ACA in terms of optimal path length (OPL), average path length (APL), and average iteration times in three simulation environments: simple, moderately complex, and complex. And in both simulation environments, the improved ant colony optimisation (ACO) performs more significantly than ACO. In moderately complex environments, the improved ACO reduces the APL and iteration times by 3.58 and 96.13, respectively. In complex environments, improving ACO reduces the APL and iteration times by 0.83 and 15.4, respectively. This model’s development and research aims to provide novel concepts for enhanced application of service robots.
Important Links:
Go Back
