ADAS Horizon Generation Using Path Probability Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced Driver Assistance Systems (ADAS) face challenges in generating a reliable horizon for predicting vehicle paths, especially when drivers deviate from pre-calculated routes, leading to potential 'blind spots' until new horizons can be generated, which affects the accuracy and reliability of ADAS functionality.
Innovation Solution
The method involves using stored digital location-based data, vehicle data, and driver data to generate an ADAS horizon, which includes predicting the relative probability of each possible path at decision points, allowing for more accurate and reliable path prediction and ADAS operation, even when the vehicle diverges from expected routes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the ADAS horizon comprises data relating only to the road currently being traversed up to a predetermined distance, then the data transmission load is reduced, but the ADAS applications are left 'blind' for a time if the driver deviates from the road currently being traversed
Solution Approach 1:
The system pre-calculates multiple possible paths the vehicle may take and includes horizon data for these alternative paths in advance. When the vehicle deviates from the current road, the ADAS applications can immediately utilize the pre-prepared horizon data for the new path without experiencing a 'blind' period, thus maintaining reliability while managing data transmission load
2Reliability
If the ADAS horizon includes multiple possible paths and alternative routes, then the system can accommodate driver deviations without 'blind spots', but the data transmission load and processing complexity increase
Solution Approach 1:
The horizon data is segmented into multiple discrete paths, each with its own set of road attributes and characteristics. This segmentation allows the system to manage and process alternative routes as separate, organized units, reducing the overall complexity of handling multiple paths while ensuring comprehensive coverage for reliable ADAS operation
3Measurement precision
If the ADAS horizon is generated frequently to account for route deviations, then path prediction accuracy is improved, but the computational resources and processing time are consumed
Solution Approach 1:
The system generates and updates horizon data periodically at predetermined intervals rather than continuously. This periodic generation maintains adequate path prediction accuracy for the duration between updates while significantly reducing computational energy consumption compared to continuous real-time regeneration of horizon data
Data Source
AI summary
A method of generating a horizon for use by an ADAS of a vehicle involves using digital location-based data, driver data and/or vehicle data to determine the likelihood that different outgoing paths are taken at a decision point along a currently traversed road segment, and deriving a probability that each path may be taken. The probability may be based on one or more of: an angle of the path relative to the incoming path, the road class of the path, a speed profile of the path, historical paths taken by vehicles at the decision point, and historical paths taken at the decision point by the individual driver or vehicle.