Autonomous Vehicle Sensor Calibration Using Native Roadside Reference Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current calibration methods for autonomous vehicle sensors are complex, inefficient, and cannot be performed in real-time during driving missions, leading to inaccuracies and reduced reliability in object detection and tracking.
Innovation Solution
A system that performs run-time calibration of autonomous vehicle sensors using native scenery encountered during driving missions, identifying reference points with multiple sensors and optimizing sensor parameters to minimize coordinate errors, allowing for fast and accurate calibration without the need for specialized facilities or off-time procedures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration methods are used, then calibration accuracy can be maintained, but the calibration process becomes complex and cannot be performed in real-time during driving missions
Solution Approach 1:
The system performs self-calibration using naturally occurring reference points in the driving environment (buildings, trees, poles, traffic signs) without requiring external calibration facilities or specialized equipment. The autonomous vehicle uses its own sensors to identify reference points, estimate their locations, and adjust sensor parameters autonomously during normal operation.
Solution Approach 2:
The calibration system uses the vehicle's existing sensors (cameras, LIDAR, radar) for both their primary sensing functions and for calibration purposes. The same sensors that detect objects for navigation are also used to identify reference points and determine their locations, eliminating the need for separate calibration equipment.
2Measurement precision
If traditional calibration methods are used, then calibration accuracy can be maintained, but calibration cannot be performed during actual driving missions requiring specialized facilities
Solution Approach 1:
The system performs self-calibration using naturally occurring reference points in the driving environment (buildings, trees, poles, traffic signs) without requiring external calibration facilities or specialized equipment. The autonomous vehicle uses its own sensors to identify reference points, estimate their locations, and adjust sensor parameters autonomously during normal operation.
Solution Approach 2:
The calibration system operates dynamically during actual driving missions rather than requiring static off-time procedures. The vehicle continuously identifies reference points and performs calibration adjustments in real-time as it moves through the environment, adapting to changing driving conditions and scenery.
3Productivity
If run-time calibration is implemented, then calibration can be performed during driving missions, but system complexity increases
Solution Approach 1:
The calibration system uses the vehicle's existing sensors (cameras, LIDAR, radar) for both their primary sensing functions and for calibration purposes. The same sensors that detect objects for navigation are also used to identify reference points and determine their locations, eliminating the need for separate calibration equipment.
Solution Approach 2:
The system performs self-calibration using naturally occurring reference points in the driving environment (buildings, trees, poles, traffic signs) without requiring external calibration facilities or specialized equipment. The autonomous vehicle uses its own sensors to identify reference points, estimate their locations, and adjust sensor parameters autonomously during normal operation.
4Measurement precision
If multiple sensors are used for calibration, then calibration accuracy improves, but computational complexity increases
Solution Approach 1:
The system uses a loss function that computes the difference between estimated locations of the same reference point obtained from different sensors. This feedback mechanism guides the optimization process to adjust sensor parameters (extrinsic and intrinsic) to minimize discrepancies, thereby improving calibration accuracy through iterative refinement.
Solution Approach 2:
The calibration process optimizes sensor parameters including extrinsic parameters (sensor positions and orientations) and intrinsic parameters (focal lengths, principal points). By adjusting these parameters to minimize the loss function, the system achieves accurate multi-sensor calibration without requiring complex manual intervention.
Data Source
AI summary
The described aspects and implementations enable efficient calibration of a sensing system of an autonomous vehicle (AV). In one implementation, disclosed is a method and a system to perform the method, the system including the sensing system configured to collect sensing data and a data processing system, operatively coupled to the sensing system. The data processing system is configured to identify reference point(s) in an environment of the AV, determine multiple estimated locations of the reference point(s), and adjust parameters of the sensing system based on a loss function representative of differences of the estimated locations.