Abstract
This paper discusses the navigation of the home service robot with least power consumption. Firstly, the localization of position of the robot with least power consumption is analyzed based on the geometrical relationship between robot and home appliance, then the localization of angle of the robot in the indoor environment with least power is proposed, which can be predicted by machine learning algorithms. Following, the path planning of the robot with least power is proposed, two supplements and optimizations of previous algorithms are proposed, the power consumption on the moving path is calculated according to the law of energy conservation, the conclusion is obtained that moving straight path and rotating in place will save more power than moving curve path. Then, the obstacle avoidance of the robot in the dynamic environment with least power consumption is proposed, the navigation process with least power consumption is analyzed and proposed in the indoor dynamic environment. Lastly, the experiment is conducted in the home to prove that compared with the DWA and TEB algorithm, there is least power consumption for the proposed navigation algorithm. The five novel points are summarized and proposed in the conclusion to save more power consumption in the localization of position and angle, path planning and navigation process of the robot.