Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Automated Parking Planning with Vision-Based BEV Approach

Yuxuan Zhao

TL;DR

An improved automated parking algorithm based on the A* algorithm is proposed, integrating vehicle kinematic models, heuristic function optimization, bidirectional search, and Bezier curve optimization to enhance the computational speed and real-time capabilities of the planning algorithm.

Abstract

Automated Valet Parking (AVP) is a crucial component of advanced autonomous driving systems, focusing on the endpoint task within the "human-vehicle interaction" process to tackle the challenges of the "last mile".The perception module of the automated parking algorithm has evolved from local perception using ultrasonic radar and global scenario precise map matching for localization to a high-level map-free Birds Eye View (BEV) perception solution.The BEV scene places higher demands on the real-time performance and safety of automated parking planning tasks. This paper proposes an improved automated parking algorithm based on the A* algorithm, integrating vehicle kinematic models, heuristic function optimization, bidirectional search, and Bezier curve optimization to enhance the computational speed and real-time capabilities of the planning algorithm.Numerical optimization methods are employed to generate the final parking trajectory, ensuring the safety of the parking path. The proposed approach is experimentally validated in the commonly used industrial CARLA-ROS joint simulation environment. Compared to traditional algorithms, this approach demonstrates reduced computation time with more challenging collision-risk test cases and improved performance in comfort metrics.

Automated Parking Planning with Vision-Based BEV Approach

TL;DR

An improved automated parking algorithm based on the A* algorithm is proposed, integrating vehicle kinematic models, heuristic function optimization, bidirectional search, and Bezier curve optimization to enhance the computational speed and real-time capabilities of the planning algorithm.

Abstract

Automated Valet Parking (AVP) is a crucial component of advanced autonomous driving systems, focusing on the endpoint task within the "human-vehicle interaction" process to tackle the challenges of the "last mile".The perception module of the automated parking algorithm has evolved from local perception using ultrasonic radar and global scenario precise map matching for localization to a high-level map-free Birds Eye View (BEV) perception solution.The BEV scene places higher demands on the real-time performance and safety of automated parking planning tasks. This paper proposes an improved automated parking algorithm based on the A* algorithm, integrating vehicle kinematic models, heuristic function optimization, bidirectional search, and Bezier curve optimization to enhance the computational speed and real-time capabilities of the planning algorithm.Numerical optimization methods are employed to generate the final parking trajectory, ensuring the safety of the parking path. The proposed approach is experimentally validated in the commonly used industrial CARLA-ROS joint simulation environment. Compared to traditional algorithms, this approach demonstrates reduced computation time with more challenging collision-risk test cases and improved performance in comfort metrics.
Paper Structure (45 sections, 19 equations, 24 figures, 6 tables)

This paper contains 45 sections, 19 equations, 24 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Characteristics of Automated Parking
  4. Traditional Planning Algorithms
  5. Methodology
  6. Overall Architecture
  7. Overall Workflow
  8. Kinematic Model
  9. Optimized Path Planning
  10. Ego Vehicle Volume
  11. Reduce Search Time
  12. Heuristic Function Optimization
  13. Bidirectional Search
  14. Binary Heap Optimization
  15. Bezier Curve Smoothing Optimization
  16. ...and 30 more sections

Figures (24)

  • Figure 1: Overall architecture of the automatic parking planning method proposed in this paper.
  • Figure 2: Vehicle Kinematic Model
  • Figure 3: Trajectories that are too close to the wall or cross corners. These trajectories obviously cannot be used as parking trajectories in the real world.
  • Figure 4: Some papers and projects use similar maps for testing. Compared to real parking maps, the obstacles in these maps are too sparse.
  • Figure 5: Grid map corresponding to a real scene
  • ...and 19 more figures