Vehicle Sensor Calibration Using Voxel Residual Minimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor calibration techniques for vehicles are time-consuming, computationally expensive, and often require offline processes, necessitating vehicles to be taken out of service, and are limited in their ability to perform calibration while the vehicle is in operation or communicating with remote servers.
Innovation Solution
A method and system for calibrating vehicle sensors, such as LIDAR sensors, using data collected during operation within a designated region of interest, which allows for online calibration and verification, utilizing a voxel space representation to minimize residual values through an anxious search algorithm, enabling simultaneous calibration of multiple sensors while the vehicle is in operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If current calibration techniques are used, then sensor alignment accuracy is improved, but vehicle downtime increases and operational continuity deteriorates
Solution Approach 1:
The calibration system transitions from static offline calibration to dynamic online calibration. The system performs calibration computations while the vehicle is operating, using real-time sensor data from the operational environment. This dynamic approach allows calibration to occur during normal vehicle use rather than requiring vehicle shutdown, thus maintaining sensor alignment accuracy while eliminating vehicle downtime.
Solution Approach 2:
The system performs preliminary calibration computations using data collected during normal vehicle operation. By continuously gathering sensor data and performing preliminary calibration calculations in the background during vehicle operation, the system prepares calibration results that can be applied without requiring vehicle shutdown, thus maintaining both accuracy and operational continuity.
2Measurement precision
If traditional calibration methods are used, then calibration accuracy is improved, but computational cost and processing time increase
Solution Approach 1:
The calibration computation is divided into multiple segments that can be processed independently and in parallel. The system segments the calibration problem into discrete computational tasks that can be distributed across multiple processing units, reducing the computational burden on any single processor while maintaining overall calibration accuracy through coordinated processing of all segments.
Solution Approach 2:
The system performs partial calibration computations during vehicle operation using available operational data, rather than requiring complete offline calibration. By performing calibration incrementally using partial datasets collected during normal operation, the system achieves sufficient calibration accuracy while significantly reducing computational cost and processing time compared to traditional comprehensive offline calibration methods.
3Manufacturing precision
If offline calibration processes are used, then thorough calibration is achieved, but vehicle service interruption increases
Solution Approach 1:
The calibration process is made continuous by performing computations during normal vehicle operation rather than requiring vehicle shutdown. The system continuously collects sensor data during vehicle operation and continuously performs calibration computations in the background, maintaining both thorough calibration and uninterrupted vehicle service through continuous useful action during operational periods.
Solution Approach 2:
The vehicle calibration system performs self-calibration using its own operational sensor data without requiring external calibration equipment or vehicle shutdown. The vehicle uses its normally-collected sensor data from operational environments to automatically perform calibration computations, achieving thorough calibration while maintaining continuous operation through self-service capability.
Data Source
AI summary
Perception sensors of a vehicle can be used for various operating functions of the vehicle. A computing device may receive sensor data from the perception sensors, and may calibrate the perception sensors using the sensor data, to enable effective operation of the vehicle. To calibrate the sensors, the computing device may project the sensor data into a voxel space, and determine a voxel score comprising an occupancy score and a residual value for each voxel. The computing device may then adjust an estimated position and/or orientation of the sensors, and associated sensor data, from at least one perception sensor to minimize the voxel score. The computing device may calibrate the sensor using the adjustments corresponding to the minimized voxel score. Additionally, the computing device may be configured to calculate an error in a position associated with the vehicle by calibrating data corresponding to a same point captured at different times.