Inferring Lane Boundaries via Vehicle Telemetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems for autonomous and semi-autonomous vehicles face challenges in determining lane boundaries, especially in areas with unmarked or confusing lane markings, intersections, roads under overpasses, and conditions like construction or snow, where sensor data is insufficient or obscured.
Innovation Solution
A system that uses high-speed vehicle telemetry to infer lane boundaries by aggregating, normalizing, and denoising vehicle trajectory data to generate unified lane geometries, which are then matched with established lanes to provide accurate navigation guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is used to determine lane boundaries, then lane markings can be directly detected, but detection fails in areas with unmarked or confusing lane markings, intersections, roads under overpasses, and conditions like construction or snow
Solution Approach 1:
The patent uses vehicle telemetry data as an intermediary to infer lane boundaries when direct sensor detection fails. Instead of relying solely on sensor data that cannot penetrate overpasses or detect unmarked roads, the system aggregates trajectory information from multiple vehicles to indirectly determine lane geometries in complex road sections
Solution Approach 2:
The system creates a virtual copy of lane boundaries by inferring them from aggregated vehicle trajectory data. When physical lane markings are absent or obscured, the patent reconstructs lane geometries by analyzing and synthesizing movement patterns of multiple vehicles, effectively copying the intended road layout from behavioral data
2Reliability
If vehicle telemetry is aggregated to infer lane boundaries, then navigation can be provided for road sections with insufficient lane markings, but data processing complexity increases
Solution Approach 1:
The patent segments the road network into different types of road sections based on lane marking quality. Road sections are categorized as having sufficient or insufficient lane markings, allowing the system to apply different processing strategies - using direct sensor detection for well-marked sections and telemetry-based inference for complex sections, thereby managing data processing complexity efficiently
Solution Approach 2:
The system applies telemetry aggregation selectively rather than universally - only for road sections identified as having insufficient or obscured lane markings. This partial action approach processes telemetry data extensively only where needed, avoiding unnecessary computational overhead in well-marked sections while maintaining navigation reliability where required
Data Source
AI summary
A system for inferring lane boundaries via vehicle telemetry for a road section is provided. The system includes a computerized remote server device operable to receive through a communications network sensor data describing lane markings upon roads bordering the road section, determine established lanes upon roads bordering the road section based upon the sensor data, receive through the communications network the vehicle telemetry generated by a plurality of vehicles traversing the road section, generate inferred lanes for the road section based upon the vehicle telemetry, match the inferred lanes to the established lanes to generate unified lane geometries, and publish the unified lane geometries.