Route Lane Matching via Graph Search for Vehicle Action Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing planning systems in autonomous and semi-autonomous vehicles struggle to accurately predict future actions of objects in the environment, leading to inefficiencies in computational resources and potential safety hazards due to inadequate consideration of road segments.
Innovation Solution
A computing device extracts points from an object trajectory and identifies a reduced set of road segments using a graph search algorithm, considering properties such as geometry and traffic rules, to enhance prediction accuracy and resource efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a planning system considers all road segments in the environment, then prediction accuracy of future actions is improved, but computational resources are wasted and processing speed decreases
Solution Approach 1:
The patent segments the complete set of road segments into two subsets: a first subset of road segments identified as relevant to the object's trajectory, and a second subset of road segments excluded from consideration. This segmentation allows the system to focus computational resources only on the relevant subset, thereby improving processing speed while maintaining prediction accuracy for the object's future actions.
Solution Approach 2:
The patent extracts and identifies a specific subset of road segments from the complete environment that are relevant to the object's trajectory. By taking out only the necessary road segments from the full set, the system reduces computational load and processing time while preserving the accuracy needed to predict future actions, thus resolving the contradiction between considering all segments and processing speed.
2Reliability
If a planning system processes all road segments, then comprehensive coverage is achieved, but computational resources are inefficiently used
Solution Approach 1:
The patent divides the road segments into a first subset that is processed for comprehensive coverage and a second subset that is excluded. This segmentation ensures that computational resources are efficiently used by processing only the necessary segments while maintaining reliable coverage of all relevant road segments for accurate prediction.
Solution Approach 2:
The patent extracts the essential subset of road segments from the complete set, achieving comprehensive coverage of relevant segments while eliminating unnecessary processing of irrelevant segments. This extraction approach maintains reliability by ensuring all necessary road segments are considered while improving computational resource efficiency.
3Productivity
If the planning system reduces the set of road segments, then computational efficiency is improved, but prediction accuracy may deteriorate
Solution Approach 1:
The patent applies local quality by identifying and processing road segments with different relevance characteristics differently. The first subset of road segments is selected based on their local relevance to the object's trajectory, ensuring that prediction accuracy is maintained for critical areas while reducing processing of less relevant segments, thus achieving both computational efficiency and accurate prediction.
Data Source
AI summary
Techniques for identifying road segments associated with an object trajectory are discussed herein. A computing device can implement a model that extracts points from an object trajectory and identifies road segments along a path between the extracted points. The model can determine a set of road segments for the path based on searching a graph. The graph can comprise nodes to represent different road segments, and a graph search algorithm can output road segments representing a shortest path between the nodes associated with the extracted points. The road segments can be used by a vehicle computing device for predicting vehicle actions to control a vehicle.