Robust Parking Path Planning with Error-Adaptive Sampling under Perception Uncertainty

Abstract

In automated parking systems, a path planner generates a path to reach the vacant parking space detected by a perception system. To generate a safe parking path, accurate detection performance is required. However, the perception system always includes perception uncertainty, such as detection errors due to sensor noise and imperfect algorithms. If the parking path planner generates the parking path under uncertainty, problems may arise that cause the vehicle to collide due to the automated parking system. To avoid these problems, it is a challenging problem to generate the parking path from the erroneous parking space. To solve this conundrum, it is important to estimate the perception uncertainty and adapt the detection error in the planning process. This paper proposes a robust parking path planning that combines an error-adaptive sampling of generating possible path candidates with a utility-based method of making an optimal decision under uncertainty. By integrating the sampling-based method and the utility-based method, the proposed algorithm continuously generates an adaptable path considering the detection errors. As a result, the proposed algorithm ensures that the vehicle is safely located in the true position and orientation of the parking space under perception uncertainty.

Data Efficient Reinforcement Learning for Integrated Lateral Planning and Control in Automated Parking System

“…The real vehicle experimental results showed that the final parking performance of DERL, which did not use system identification, was better than that of MCTS guided by a policy network using a refined vehicle model, which planed fewer motions and was more affected by vehicle model errors and perception errors. Similar to [1,8,19], re-planning proved to be a powerful tool for autonomous driving systems under perception uncertainty.…”

“…The former includes numerical optimizations, A* search, rapidly exploring random tree (RRT) family, and curve planners. In the optimization methods, local optimization [ 1 , 19 ] was developed to find the short-range path. Although the global optimality cannot be guaranteed, they usually generated feasible paths in a short time.…”

“…The LiDAR can detect the parking space (i.e., the space between two vehicles) [ 5 ]. After the detection of the parking space, the vehicle pose was estimated by dead reckoning, using the data of the motion sensors [ 19 , 24 ].…”

A GRASP-Based Approach for Planning UAV-Assisted Search and Rescue Missions

“…Another important aspect is safety, as obstacles such as geographic obstructions, housing, or uncertainty in GPS precision and speed should be considered. In the works of Lee et al [ 19 ] an optimal path was found while considering collision while, e.g., Sung et al [ 20 ] consider online updating. This is a field on its own as this needs a discussion of online vs. offline planning.…”

Path Planning in Localization Uncertaining Environment Based on Dijkstra Method

“…Papachristos et al ( 2019 ) designed a paradigm that follows a hierarchical optimization objective and executes it in a backward horizon manner to implement an uncertainty-aware path planning strategy. Combining adaptive error sampling for generating possible path candidates with a utility-based approach, Lee et al ( 2020 ) implements a path planning task for safe parking under perceptible uncertainty, which takes into account detection errors and makes optimal decisions under uncertainty. Uncertainty generation is mainly obtained through passive sensors, and unfortunately, the current capability to rely on inertial navigation alone for path planning under uncertainty needs to be further explored.…”