Autonomous Vehicle Route Planning at Road Junctions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional route planning algorithms for autonomous driving vehicles often result in unexpected navigation routes when the vehicle is located at a road junction, due to difficulties in finding a correct start search node, leading to incorrect and inefficient routes.
Innovation Solution
A method that generates a Cartesian product of sets of topology nodes for all candidate navigation routes, using the A-star search algorithm to traverse and merge these routes, and selecting the best candidate route based on cost, ensuring accurate and efficient navigation through road junctions and regular scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional A-star searching algorithm is used for route planning, then the algorithm is simple to implement, but it fails to find correct routes when start or destination points are located at road junctions
Solution Approach 1:
The patent segments the route planning process into two distinct phases: (1) generating multiple candidate routes using improved A-star algorithm with road junction detection, and (2) selecting the optimal route using cost function evaluation. This segmentation allows each phase to specialize, improving overall reliability without overwhelming complexity.
Solution Approach 2:
The patent performs preliminary detection of road junctions before executing the main route search. By identifying road junction locations in advance and using them as special search nodes, the algorithm prepares the search space appropriately, ensuring correct route finding at complex locations like road junctions.
2Reliability
If multiple candidate routes are generated and evaluated, then the optimal route can be selected with higher accuracy, but the computational time and complexity increase
Solution Approach 1:
The patent generates multiple candidate routes (excessive action) but evaluates them using an efficient cost function that prioritizes key factors like distance and road type. This allows the system to explore more options than strictly necessary while maintaining fast evaluation, achieving high accuracy without excessive computational time.
Solution Approach 2:
The patent changes the evaluation parameters by introducing a comprehensive cost function that considers multiple factors (distance, road type, traffic conditions) rather than simple distance metrics. This parameter transformation enables better route selection accuracy while maintaining computational efficiency through structured evaluation.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Disclosed herein is a computer-implemented method for operating an autonomous driving vehicle (ADV), comprising: determining a starting point, a set of one or more way points, and a destination point of a first route along which the ADV is to be driven(1001);determining all lane segments near the starting point, the set of way points, and the destination point within a predetermined threshold distance respectively(1002); determining a set of route candidates using an A-star (A*) searching algorithm based on a set of nodes representing all lane segments near the starting point, the set of way points, and the destination point respectively(1003); selecting a second route from the set of route candidates based on respective costs of the set of route candidates(1004); and controlling the ADV to drive along the selected route autonomously(1005).