Autonomous Vehicle Agent Prioritization for Real-Time Trajectory Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in generating timely and accurate prediction data for agents in their vicinity due to limited computational resources, especially in crowded environments, which can impact safe and comfortable trajectory planning.
Innovation Solution
The system prioritizes agents based on importance scores generated by an importance scoring model, concentrating computational resources on high-priority agents and using less computationally intensive models for low-priority agents, representing trajectories as 2D channels to optimize processing efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prediction data is generated for all agents in the vicinity using a comprehensive prediction model, then the accuracy and reliability of trajectory planning is improved, but the computational resource consumption and processing time increase significantly
Solution Approach 1:
The patent segments the set of all agents into high-priority and low-priority subsets based on importance scores. High-priority agents (those with higher importance scores) are processed using a comprehensive prediction model to ensure accurate trajectory planning, while low-priority agents are handled with simplified models or lower processing frequency. This segmentation allows the system to allocate computational resources selectively, maintaining safety for critical agents while reducing overall computational burden.
2Measurement precision
If a comprehensive prediction model is used for all agents, then the accuracy of prediction data is improved, but the processing speed and real-time performance deteriorate
Solution Approach 1:
The patent applies local quality by using different prediction model complexities for different agents based on their importance scores. High-priority agents receive detailed, accurate predictions using comprehensive models, while low-priority agents are processed with simpler, faster models. This differential approach ensures that computational accuracy is concentrated where it is most needed for safe trajectory planning, while overall processing speed is maintained through efficient handling of less critical agents.
3Loss of information
If computational resources are allocated to process all agents equally, then the completeness of prediction data is improved, but the system responsiveness and timeliness deteriorate
Solution Approach 1:
The patent implements dynamic resource allocation by continuously calculating importance scores for agents and adjusting the level of processing applied to each agent in real-time. Agents are dynamically reclassified as high or low priority based on their current importance scores, which may change as the vehicle's trajectory and the agents' positions evolve. This dynamic approach ensures that computational resources are always allocated to the most relevant agents, maintaining prediction completeness for critical cases while minimizing processing time overall.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying high-priority agents in the vicinity of a vehicle. In one aspect, a method comprises processing an input that characterizes a trajectory of the vehicle in an environment using an importance scoring model to generate an output that defines a respective importance score for each of a plurality of agents in the environment in the vicinity of the vehicle. The importance score for an agent characterizes an estimated impact of the agent on planning decisions generated by a planning system of the vehicle which plans a future trajectory of the vehicle. The high-priority agents are identified as a proper subset of the plurality of agents with the highest importance scores.