Road-Geofenced Vehicle Positioning With Dead-Reckoning Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle positioning systems, particularly in V2X applications, face inaccuracies due to satellite signal errors like multipath effects in urban areas, leading to significant positioning and heading angle errors, which are exacerbated by the integration of dead-reckoning techniques without accurate initialization.
Innovation Solution
A geofencing framework is introduced that utilizes road geometry information and supplemental kinematic filters to correct vehicle positioning, incorporating timely and predicted geofencing strategies, yaw rate bias removal, and retrospective integrations to achieve road-level accuracy, thereby overcoming the limitations of GNSS and dead-reckoning errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If GNSS is used for vehicle positioning, then positioning coverage is improved, but positioning precision deteriorates due to multipath effects and atmospheric errors
Solution Approach 1:
The patent introduces road geometry information as an intermediary constraint to mediate between GNSS measurements and final positioning. The vehicle position is constrained to lie on or near the road geometry, which acts as a mediator that filters out GNSS errors while preserving coverage. This is implemented through map matching algorithms that project GNSS positions onto the road network.
Solution Approach 2:
The system implements feedback by continuously comparing the vehicle's estimated position (from GNSS and dead-reckoning) with the expected position on road geometry. When deviations are detected, the system adjusts the position estimate to conform to road constraints, creating a closed-loop correction mechanism that maintains precision without sacrificing coverage.
2Measurement precision
If dead-reckoning is integrated to improve positioning accuracy, then short-term precision is improved, but long-term reliability deteriorates due to error accumulation
Solution Approach 1:
The system performs preliminary initialization of dead-reckoning using GNSS positions that have been pre-constrained to road geometry. By establishing the initial position and orientation on the road network before dead-reckoning begins, the system prevents error accumulation from poor initialization. The road geometry serves as a preliminary reference frame for the dead-reckoning process.
Solution Approach 2:
The system continuously monitors dead-reckoning position estimates against road geometry constraints and provides feedback corrections. When the dead-reckoning position deviates from the road network beyond acceptable thresholds, the system resets or corrects the position to the nearest valid road location, preventing unbounded error accumulation and maintaining long-term reliability.
3Measurement precision
If geofencing framework is implemented to achieve road-level accuracy, then positioning precision is improved, but system complexity increases
Solution Approach 1:
The road geometry data structure serves multiple functions simultaneously: it defines the valid positioning area (geofence), provides constraints for map matching, serves as a reference for dead-reckoning initialization, and acts as a correction reference for error compensation. This multi-functionality reduces the need for separate geofencing data structures and processing logic, managing complexity while achieving road-level accuracy.
Data Source
AI summary
This provides methods and systems for the global navigation satellite system (GNSS) combined with the dead-reckoning (DR) technique, which is expected to provide a vehicle positioning solution, but it may contain an unacceptable amount of error due to multiple causes, e.g., atmospheric effects, clock timing, and multipath effect. Particularly, the multipath effect is a major issue in the urban canyons. This invention overcomes these and other issues in the DR solution by a geofencing framework based on road geometry information and multiple supplemental kinematic filters. It guarantees a road-level accuracy and enables certain V2X applications which does not require sub-meter accuracy, e.g., signal phase timing, intersection movement assist, curve speed warning, reduced speed zone warning, and red-light violation warning. Automated vehicle is another use case. This is used for autonomous cars and vehicle safety, shown with various examples/variations.