Vehicle pose estimation algorithm for parking automated guided vehicle

Abstract

Parking automated guided vehicle is more and more widely used for efficient automatic parking and one of the tough challenges for parking automated guided vehicle is the problem of vehicle pose estimation. The traditional algorithms rely on the profile information of vehicle body and sensors are required to be mounted at the top of the vehicle. However, the sensors are always mounted at a lower place because the height of a parking automated guided vehicle is always beyond 0.2mm, where we can only get the vehicle wheel information and limited vehicle body information. In this article, a novel method is given based on the symmetry of wheel point clouds collected by 3-D lidar. Firstly, we combine cell-based method with support vector machine classifier to segment ground point clouds. Secondly, wheel point clouds are segmented from obstacle point clouds and their symmetry are corrected by iterative closest point algorithm. Then, we estimate the vehicle pose by the symmetry plane of wheel point clouds. Finally, we compare our method with registration method that combines sample consensus initial alignment algorithm and iterative closest point algorithm. The experiments have been carried out.

Trajectory Tracking Control of Parking Automated Guided Vehicles Using Nonlinear Disturbance Observer-Based Sliding Mode

“…However, the difference is that its comb-arms can pass through the comb rack to lift the tires, and the comb racks will not be carried when transporting the vehicles [14] . Recently, a new clamping P-AGV has been developed, whose tire clamping arms can automatically clamp and lift the tires without the assistance of any other devices [15,16] .…”

“…On the basis of these P-AGV chassis, researchers have done a considerable amount of studies on parking space allocation, path planning and localization [16][17][18] . However, far too little attention has been paid to trajectory tracking control of the P-AGV.…”

Lane line recognition based on improved 2D-gamma function and variable threshold Canny algorithm under complex environment

“…Compared with particle filter or Kalman filter, the tracking failure rate is lower, but it can’t deal with the problems such as the change of light rays and the incomplete identification lines; 6 Xing and sun et al Proposed a light adaptive image segmentation method based on relax constraint support vector machine (SVM) classifier and kernel function to distinguish the original color features and light artifacts of road image, and achieved good results. However, this study did not focus on the situation that the color of the sign belt is close to the ground, nor did it consider the possible disordered lines on the ground Interference; 7 The team also carried out specific research on path recognition, and proposed an intelligent path recognition method based on KPCA-BPNN and IPSO-BTGWP to resist the interference of complex workspace, while the algorithm is complex and difficult to be directly applied to AGV; 8 Ning proposed a new method based on the symmetry of 3D lidar wheel point cloud to estimate the vehicle position and attitude, which solved the problem of limited body attitude information caused by the low installation position of AGV sensor; 9 Yang Yanyu corrects the single visual perception by building the GPS curve model of preview model and target track based on the GPS data, and completes the independent tracking. However, this method needs to manually make the vehicle complete the data collection according to the expected track before the vehicle completely runs autonomously, which is not widely applicable.…”