Mutual Agent Ranking for Autonomous Vehicle Prediction Load
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 where many agents need to be considered, leading to insufficient processing capacity for all agents.
Innovation Solution
The system determines mutual importance scores for agents by assessing their relevance to the vehicle's and each other's planning decisions, concentrating computational resources on high-priority agents and using less intensive models for others, allowing for precise prediction data generation for critical agents while reducing resource usage on less impactful ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computational resources are allocated to generate prediction data for all agents in crowded environments, then measurement precision of agent interactions is improved, but productivity of the system deteriorates due to insufficient processing capacity
Solution Approach 1:
The patent applies local quality by differentiating the level of analysis between different agents based on their mutual importance scores. High-priority agents receive comprehensive prediction data generation using intensive computational models, while low-priority agents receive simplified or no prediction analysis. This selective application of computational resources maintains measurement precision for critical agents while improving overall system productivity by reducing unnecessary processing.
2Reliability
If comprehensive prediction data is generated for all agents, then reliability of planning decisions is improved, but loss of time increases due to extended processing duration
Solution Approach 1:
The patent extracts only the essential prediction data needed for reliable planning decisions by identifying and processing only high-priority agents whose actions significantly impact vehicle planning. Less critical agents are excluded from intensive prediction analysis. This extraction approach maintains reliability for decision-critical agents while reducing overall processing time by eliminating unnecessary computations for low-impact agents.
3Measurement precision
If intensive importance scoring models are used for all agents, then measurement precision of agent relevance is improved, but use of energy increases beyond available computational resources
Solution Approach 1:
The patent applies local quality by varying the computational intensity of importance scoring models based on agent priority levels. High-priority agents undergo comprehensive scoring analysis using intensive models to ensure measurement precision, while low-priority agents receive simplified scoring or are excluded from detailed analysis. This selective approach maintains energy efficiency by concentrating computational resources only where high precision is necessary for safe operation.
Data Source
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. The high-priority agents can be identified based on a set of mutual importance scores in which each mutual importance score indicates an estimated mutual relevance between the vehicle and a different agent from a set of agents on planning decisions of the other. The mutual importance scores can be calculated based on importance scores assessed from the perspectives of both the vehicle and the agents.