Autonomous Vehicle Roadway Model Discrepancy Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving road vehicles face navigation challenges and performance issues due to outdated roadway models, which require frequent and costly updates of the entire network, making it inefficient to maintain currency of the model.
Innovation Solution
A control system with an algorithm that determines discrepancy data from self-driving road vehicles, aggregates it, and analyzes it to identify areas needing updates, balancing the need for model currency with cost considerations, using sensor data from various sources like image, radar, and LIDAR to pinpoint necessary updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire roadway network is traversed to update the roadway model, then the model currency is improved, but the cost and time required for updating increases significantly
Solution Approach 1:
The patent divides the roadway network into discrete segments and uses multiple self-driving vehicles to independently traverse and collect data from different segments. This segmentation allows parallel data collection across the network, reducing the total time and cost required to update the roadway model while maintaining comprehensive coverage.
Solution Approach 2:
Instead of requiring complete traversal of the entire roadway network, the patent accepts partial data from multiple vehicles that collectively cover sufficient portions of the network. The system processes this partial data to identify specific areas needing updates, avoiding the excessive cost of comprehensive full-network traversal while still achieving adequate model currency.
2Reliability
If the entire roadway network is traversed to update the roadway model, then the model currency is improved, but the time required for updating increases
Solution Approach 1:
The roadway network is divided into segments that can be traversed simultaneously by multiple self-driving vehicles. This parallel processing approach reduces the total time required to collect sufficient data for roadway model updates, as vehicles operate independently across different segments at the same time.
Solution Approach 2:
Self-driving vehicles autonomously traverse roadways and collect their own sensor data without requiring human intervention or coordination. This self-service capability enables continuous, unsupervised data collection that significantly reduces the time required to gather sufficient information for roadway model updates.
3Measurement precision
If discrepancy data from multiple vehicles is aggregated, then the accuracy of identifying update areas is improved, but the data processing complexity increases
Solution Approach 1:
The system extracts and isolates only the discrepancy data from sensor readings that differ from the existing roadway model, rather than processing all sensor data. This extraction approach focuses computational resources on identifying specific areas needing updates, reducing overall data processing complexity while maintaining high accuracy in detecting roadway changes.
Solution Approach 2:
A centralized server acts as an intermediary that receives, aggregates, and processes discrepancy data from multiple self-driving vehicles. This intermediary consolidates the complex task of comparing and analyzing data from numerous vehicles into a single processing point, simplifying the overall system architecture while enabling accurate identification of roadway areas requiring model updates.
Data Source
AI summary
The present disclosure relates to an autonomous vehicle and methods of operating an autonomous vehicle. A sensor for the autonomous vehicle generates sensor data for a current location of the autonomous vehicle. A positioning system generates data for identifying the current location of the vehicle. A computing system of the autonomous vehicle identifies a portion of a roadway model which corresponds to the current location of the vehicle. The roadway model has a level of detail suitable for autonomous operation. The computing system identifies whether a discrepancy is present between the identified portion of the roadway model and the sensor data. When the computing system identifies the discrepancy, the computing system communicates data which corresponds to the discrepancy to a roadway model management system configured to update the roadway model.