Design of robot navigation system under spatial constraints in shiitake mushroom fruiting room

From the paper

Abstract

In response to the challenges faced by traditional robot navigation solutions in shiitake mushroom fruiting room, such as narrow aisles, poor satellite signal, sparse spatial structure and incomplete sensor data collection, a robot navigation system was proposed under spatial constraints in mushroom fruiting room. First, according to the issue of noise and uncertainty in the data collected by odometers and inertial measurement unit (IMU) sensors, an error-state Kalman filter (ESKF) was used to fuse both sensor data to improve positioning accuracy. Then, a dual LiDAR data fusion algorithm was proposed to integrate environmental information at different heights. Subsequently, an improved Cartographer LiDAR mapping algorithm was used to construct a navigation grid map. During the navigation process, based on Navigation2 autonomous navigation framework, the global localization was achieved by using the adaptive Monte Carlo localization algorithm, the Theta* algorithm was applied as the planner algorithm for the planning server, while the control server used a dynamic window-based local path planning algorithm to control robot movement. The experimental results in shiitake mushroom fruiting room showed that the obstacle coverage in the grid map constructed through the dual LiDAR fusion algorithm was improved by 2.07 % compared to a single LiDAR. In localization tests, when the robot moved at a speed of 0.40 m/s after fusing odometer and IMU data, the maximum values of average longitudinal deviation, average lateral deviation and average angular deviation were 5.80 cm, 3.50 cm and 3.00° respectively. The standard deviations were less than 1.47 cm, 1.17 cm and 1.16°. In navigation tests, when the robot moved at speeds of 0.20 ∼ 0.70 m/s, the average values of longitudinal deviation, lateral deviation and heading deviation between the actual path and the target path were less than 5.78 cm, 3.80 cm and 4.00° respectively, with standard deviations of less than 1.63 cm, 1.32 cm and 0.84°. The positioning and navigation accuracy can meet the requirements for robot operation in shiitake mushroom fruiting room, providing crucial technical support for the intelligent development of the edible fungus industry.