Autonomous Parking Goal Selection for Inaccessible Delivery Stops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in complex parking environments where the final parking goal is inaccessible, leading to potential vehicle stalling and traffic blockages, as existing path planning algorithms like Hybrid A* may not guarantee convergence to the optimal solution or handle error recovery effectively.
Innovation Solution
An improved system for autonomous vehicles that selects an intermediate parking goal by using search algorithms like A* and Field D* for path planning, and when the final parking goal is inaccessible, it switches to a parking plan, determining an intermediate goal through candidate node search and cost calculation, allowing the vehicle to park temporarily and monitor for accessibility of the final goal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous vehicle uses Hybrid A* algorithm for path planning, then the algorithm considers continuous search space and implements vehicle kinematics, but the algorithm cannot guarantee convergence to the optimal solution when the goal is inaccessible or not reachable
Solution Approach 1:
The patent divides the path planning process into two distinct phases: a driving plan phase for navigation to the final parking goal, and a parking plan phase for actual parking maneuvers. This segmentation allows the system to handle different stages with appropriate algorithms, improving reliability when the goal becomes inaccessible during driving but not during parking execution.
Solution Approach 2:
The patent introduces an intermediate goal point as a mediator between the current vehicle position and the final parking goal. When the final goal is inaccessible, the system calculates an intermediate point that is reachable, allowing the vehicle to park temporarily and later resume toward the original goal. This intermediary approach ensures continuous progress and prevents complete failure.
2Productivity
If the autonomous vehicle attempts to park directly at the final parking goal, then the vehicle may complete the delivery task efficiently, but the vehicle may get stuck on the road when the goal location is inaccessible or occupied
Solution Approach 1:
The system performs preliminary checks to determine whether the final parking goal is accessible before attempting direct parking. By evaluating goal accessibility in advance, the system can proactively select alternative intermediate goals, preventing the vehicle from getting stuck and ensuring reliable task completion.
Solution Approach 2:
The patent dynamically changes the parking goal parameter based on real-time accessibility conditions. When the final goal is found to be inaccessible, the system modifies the goal parameter to an intermediate point, allowing the vehicle to adapt its parking strategy and maintain high reliability in dynamic environments.
3Speed
If the autonomous vehicle uses search algorithms like A* and Field D* for path planning, then the algorithms yield fast convergence for discrete state spaces, but they cannot handle continuous coordinates and non-holonomic requirements
Solution Approach 1:
The patent segments the path planning problem into discrete path finding (using A* or Field D* for speed) and continuous parking execution (using Hybrid A* for kinematic constraints). This division allows the system to leverage the speed of discrete algorithms for navigation while using continuous algorithms for precise parking maneuvers, achieving both fast convergence and adaptability.
Data Source
AI summary
A system and a method for determining an intermediate parking goal of an autonomous vehicle. The autonomous vehicle includes a controller, the controller has a processor and a storage device storing computer executable code. The computer executable code, when executed at the processor, is configured to: prepare a driving plan toward a final parking goal and a parking plan to park at the final parking goal; drive the autonomous vehicle toward the final parking goal according to the driving plan; switch from the driving plan to the parking plan when the autonomous vehicle is within a predetermined distance to the final parking goal; and when the final parking goal is inaccessible: determine the intermediate parking goal, drive and park the autonomous vehicle at the intermediate parking goal.