Collaborative Sensor Calibration Using Route Landmarks for Trucks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving technologies do not adequately address the unique challenges of trucks and truck fleets, such as the need for advanced sensor calibration and precise monitoring of trailer positions, which are critical for safe navigation and obstacle avoidance.
Innovation Solution
A method and system for collaborative sensor calibration on-the-fly using a fleet management center and communication platform, where an ego vehicle can request assistance from other vehicles or landmarks for sensor recalibration, and utilize fiducial markers and feature-based estimation to monitor trailer positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional autonomous driving technologies are used for trucks, then general obstacle detection is possible, but sensor calibration accuracy and trailer position monitoring are insufficient
Solution Approach 1:
The patent introduces fiducial markers as intermediary objects placed on trailers to enable precise measurement. These markers serve as mediators between the camera sensor and the trailer, allowing the system to accurately track trailer position and orientation without requiring complex direct sensing. The markers provide known reference points that facilitate reliable calibration and monitoring.
Solution Approach 2:
The system creates a virtual copy of the trailer's position and orientation by detecting fiducial markers and computing trailer pose through image processing. This virtual representation allows the autonomous truck to monitor trailer position accurately without physically measuring it, enabling precise calibration and monitoring through computational methods.
2Measurement precision
If sensor calibration is performed while the truck is stationary, then calibration accuracy is maintained, but productivity and operational efficiency are reduced
Solution Approach 1:
The patent transforms the calibration process from a static operation to a dynamic one. The system performs sensor calibration while the truck is in motion by continuously tracking fiducial markers on trailers. This dynamic calibration approach maintains measurement precision while eliminating the need to stop operations, thereby preserving productivity and operational efficiency.
Solution Approach 2:
The calibration process becomes continuous rather than periodic. By using fiducial markers that are constantly visible during operation, the system maintains continuous calibration data, ensuring measurement precision is preserved throughout operations without interrupting workflow. The useful action of calibration continues uninterrupted alongside truck operations.
3Productivity
If the truck travels at higher speeds to maintain productivity, then operational efficiency improves, but the time available for calibration and monitoring decreases
Solution Approach 1:
The patent replaces mechanical calibration methods with optical and computational approaches. Instead of using physical calibration targets requiring slow, deliberate positioning, the system uses fiducial markers detected by cameras and processed through computer vision algorithms. This substitution enables calibration to occur at higher speeds, maintaining productivity while eliminating time loss.
Solution Approach 2:
The system performs preliminary computations by pre-storing fiducial marker patterns and their expected positions. When detecting markers during high-speed operation, the system compares real-time detections against pre-computed reference data, enabling rapid calibration decisions without time-consuming real-time calculations. This preliminary preparation eliminates computational delays during high-speed operation.
Data Source
AI summary
The present teaching relates to method, system, medium, and implementations for sensor calibration. An ego vehicle determines whether a sensor deployed thereon to facilitate autonomous driving needs to be calibrated and sends, if it is determined that the sensor needs to be calibrated, a request for assistance in collaborative calibration of the sensor to a center. The request includes at least a first position of the ego vehicle on the route and a first configuration of the sensor with respect to the ego vehicle. When the ego vehicle receives a calibration assistant package, which includes information associated with a collaborative means present along the route and to be used to assist the ego vehicle to calibrate the sensor, it identifies, based on the information, the collaborative means along the route when the ego vehicle is in a vicinity of the collaborative means in order to capture a target present on the collaborative means to enable calibration of the sensor.