Vehicle Sensor Positioning from Straight-Lane Offset Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor systems in vehicles with unknown positions relative to the vehicle itself cannot accurately derive trajectories due to unknown positioning, leading to degraded accuracy in map encoding, behavior planning, and machine learning model training.
Innovation Solution
Techniques to determine the lateral, longitudinal, and vertical positioning of vehicle-based sensor systems relative to the vehicle using sensor data and map information, enabling precise trajectory derivation for applications requiring high precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If sensor systems with unknown positions relative to the vehicle are used, then cost is reduced and availability is improved, but trajectory accuracy and map encoding precision deteriorate
Solution Approach 1:
The sensor system automatically determines its own positioning relative to the vehicle by analyzing sensor data collected during straight lane driving. The system identifies times when the vehicle was driving in substantially-straight lanes, calculates lateral offsets based on lane geometry and sensor measurements, and uses these calculations to self-calibrate its position without requiring external intervention or expensive pre-installation positioning equipment.
Solution Approach 2:
The system changes the operational parameters by selectively processing sensor data only during periods when the vehicle is driving in substantially-straight lanes. By filtering for these specific conditions and calculating lateral offsets based on known lane geometry during these periods, the system transforms unreliable general-purpose sensor data into precise positioning information that can be used to correct trajectory accuracy.
2Device complexity
If sensor systems with unknown positions are used, then device complexity is reduced, but map encoding accuracy and behavior planning precision deteriorate
Solution Approach 1:
The sensor system performs self-positioning by automatically analyzing its own sensor data collected during straight lane driving. It identifies relevant time periods, calculates lateral offsets based on lane geometry, and determines its position relative to the vehicle without requiring complex pre-installation procedures or additional positioning hardware, thereby maintaining simplicity while achieving high map encoding accuracy.
Solution Approach 2:
The system performs preliminary positioning calibration by collecting and analyzing sensor data during straight lane driving periods before using the sensor system for its primary functions. This preliminary action of determining lateral offset during controlled driving conditions prepares the system for accurate trajectory derivation, map encoding, and behavior planning in subsequent operations.
3Ease of operation
If sensor systems with unknown positions are used, then ease of installation is improved, but trajectory derivation precision deteriorates
Solution Approach 1:
The sensor system automatically determines its installation position relative to the vehicle by processing sensor data collected during straight lane driving. It identifies times when the vehicle was driving in substantially-straight lanes, calculates lateral offsets based on lane geometry and sensor measurements, and uses these calculations to self-calibrate, eliminating the need for complex installation procedures or professional positioning equipment.
Solution Approach 2:
The system changes the operational approach by filtering sensor data to use only measurements taken during substantially-straight lane driving periods. By calculating lateral offsets based on known lane geometry during these specific conditions and then applying these offsets to correct trajectory data, the system transforms imprecise general-purpose sensor measurements into accurate trajectory information despite simple installation.
Data Source
AI summary
Examples disclosed herein may involve a computing system that is operable to (i) identify, within a given period of operation of a vehicle having an associated sensor system for capturing sensor data, one or more times when the vehicle was driving in a lane having substantially-straight lane geometry, (ii) for each identified time, determine a respective measure of a lateral offset between the vehicle's associated sensor system and a lateral reference point of the vehicle, and (iii) based on the respective measure of the lateral offset that is determined for each of the one or more identified times, determine the lateral offset between the vehicle's associated sensor system and the lateral reference point of the vehicle.