Vehicle Path Planning With Unified Mixed-Obstacle Decision-Making
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic driving systems face difficulties in performing obstacle decision-making quickly and conveniently, especially when dealing with mixed types of obstacles, such as grid and convex hull obstacles, which complicates path planning and collision avoidance.
Innovation Solution
A vehicle decision-making planning method that generates a base coordinate system, converts grid obstacles into convex hull obstacles, and preprocesses obstacle data to reduce calculation load, allowing for unified decision-making on mixed types of obstacles, including static and dynamic obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If grid obstacles are processed directly for decision-making, then obstacle detection coverage is improved, but decision-making speed and convenience deteriorate due to high dispersion and large number of grid obstacles
Solution Approach 1:
The patent merges grid obstacles with convex hull obstacles into a unified obstacle representation system. Grid obstacles provide comprehensive coverage while convex hull obstacles provide semantic information. By combining both types and processing them through a unified decision-making framework, the system achieves both high detection coverage and fast decision-making speed.
Solution Approach 2:
The patent performs preliminary processing of grid obstacles by converting them into convex hull representations before decision-making. This preprocessing step reduces the computational burden by transforming dispersed grid data into consolidated convex hull forms, enabling faster subsequent decision-making while preserving obstacle information.
2Loss of information
If mixed type obstacles (grid and convex hull) are processed separately, then obstacle information completeness is improved, but decision-making complexity increases
Solution Approach 1:
The patent creates a universal decision-making framework that handles both grid obstacles and convex hull obstacles through the same processing pipeline. By making the decision-making module multi-functional to accommodate different obstacle types uniformly, the system maintains complete obstacle information while avoiding the complexity of separate processing paths.
Solution Approach 2:
The patent converts grid obstacles into convex hull representations as a preliminary step, unifying the obstacle data structure before decision-making. This preprocessing ensures that both grid and convex hull obstacles are in a compatible format, reducing decision-making complexity while preserving the semantic information from convex hulls and the comprehensive coverage from grids.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A vehicle decision-making planning method and apparatus, a device and a medium. The method comprises: generating a base coordinate system; generating a guide line in the base coordinate system to determine an approximate travel trajectory of a vehicle in the future; performing obstacle decision-making under the constraint of the guide line; and generating a drivable area based on the obstacle decision-making. The method can implement unified decision-making of grid type and convex hull type obstacles, simplify an obstacle decision-making process of mixed type obstacles, accelerate the obstacle decision-making process, and enable a decision-making planning module to perform obstacle decision-making conveniently and quickly.