Parking Trajectory Replanning for Dynamic Parallel Parking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking trajectory planning systems struggle to dynamically adapt to changes in the parking environment during automated parking processes, leading to invalid trajectories and process abortions, while also having high computational complexity and low user acceptance.
Innovation Solution
A method for replanning a parking trajectory that involves multiple planning strategies, including lengthening or shortening S-shaped movements and planning forward/backward parking movements to safely position the vehicle, with verification of admissible trajectories within specified limits, allowing for flexible and natural trajectory replanning with limited computational cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generic trajectory planners are used to deal with changes in detections of the surroundings, then adaptability to changing parking situations is improved, but computational complexity increases and user acceptance decreases
Solution Approach 1:
The parking trajectory planning is segmented into multiple discrete movement types (S-shaped parking movement, individual parking movements, forward/backward parking movements). Each segment type has its own planning method with specific characteristics, allowing the system to handle different situations with appropriate simplified models rather than using a single complex generic planner.
Solution Approach 2:
The patent changes parameters such as the shape of parking movements (S-shaped, circular arc, clothoid), the number and type of movements, and the planning constraints based on the detected parking situation. This allows adaptation to environmental changes while using computationally efficient parameterized movement models rather than complex generic optimization.
2Device complexity
If rule-based or geometric planning methods are used for specific parking situations, then computational complexity is reduced, but adaptability to changing environmental conditions deteriorates
Solution Approach 1:
The planning system is designed to be dynamic by allowing switching between different planning methods (first planning method for S-shaped movements, second planning method for individual movements) based on the current parking situation. The system can replan trajectories dynamically when environmental changes are detected, maintaining low computational complexity through structured decision logic.
Solution Approach 2:
The system incorporates feedback from environmental detection (sensor data about surrounding objects, parking space dimensions) to adjust the planning parameters and select appropriate planning methods. This feedback mechanism enables adaptation to changing conditions while maintaining computational efficiency through rule-based decision making.
Data Source
AI summary
A method is disclosed for replanning a parking trajectory of a vehicle during an at least semiautomatic parallel parking process following the reception of information that a pre-planned parking trajectory for the parallel parking process has to be replanned.