Road Marker Referencing for Long-Range On-Road Object Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infrastructure-based perception systems face challenges in accurately detecting the location, velocity, and acceleration of on-road objects due to limitations in sensor range and accuracy, such as limited range of LiDAR and radar sensors and increased error in distance estimation by stereo cameras as distance increases.
Innovation Solution
The use of specialized markers on or near the road, such as solid lines, patterned lines, shapes, and poles, which are detected in sensor data to determine the location of objects relative to the markers and subsequently their absolute location, velocity, and acceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If LiDAR and radar sensors are used for detection, then detection range is extended, but sensor accuracy and reliability deteriorate at longer distances
Solution Approach 1:
The patent introduces markers as intermediary objects placed on the road at known locations. These markers serve as reference points that mediate between the sensor and distant objects, enabling accurate distance and position calculations through geometric relationships rather than direct sensor measurement, thus resolving the accuracy deterioration at long distances
Solution Approach 2:
The patent transitions from direct 1D distance measurement by sensors to 2D/3D geometric positioning using markers with known coordinates. By incorporating spatial dimensions of marker positions and using triangulation/geometric calculations, the system achieves higher precision in determining object locations beyond the limitations of single-sensor range
2Reliability
If stereo cameras are used for distance estimation, then detection capability is improved, but error increases quadratically with distance
Solution Approach 1:
Markers act as intermediary reference points that provide known spatial coordinates. Instead of relying on quadratic distance estimation from stereo cameras alone, the system uses markers to establish geometric baselines, transforming the problem into one of angular measurement and triangulation, which maintains precision at longer distances
Solution Approach 2:
The patent replaces the mechanical/optical distance estimation mechanism of stereo cameras with a geometric calculation system based on known marker positions. By substituting direct distance measurement with coordinate-based geometric relationships, the system eliminates the quadratic error accumulation inherent in stereo vision
3Measurement precision
If markers are detected and used for location determination, then object location accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-placing markers at known, predetermined locations on the road before operation. The marker positions are stored in advance in the system's database, eliminating the need for real-time marker localization and reducing computational complexity during actual object detection while maintaining high location accuracy
Data Source
AI summary
Methods, systems, and non-transitory computer-readable media are configured to perform operations comprising detecting an object in sensor data captured in an environment; detecting a marker in the sensor data captured in the environment; and determining a first location of the object in the environment based on a relative location of the object to the marker and a second location of the marker.