UAV-Based Sensor Calibration for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor calibration methods for autonomous vehicles are time-consuming and spatially constrained, requiring multiple output values from variously inclined checkerboards and specific indoor spaces, and are affected by dynamic vehicle characteristics and sensor fixing device physical characteristics, necessitating frequent recalibration.
Innovation Solution
An autonomous driving device and sensor calibration system utilizing an unmanned aerial vehicle (UAV) equipped with a checkerboard, which hovers at designated waypoints and changes posture angles to generate images for sensor calibration, allowing for accurate calibration of sensors based on matching points between images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a checkerboard is placed at various inclinations throughout a visible range of a vehicle sensor for accurate calibration, then measurement precision is improved, but loss of time increases and ease of operation deteriorates
Solution Approach 1:
The patent applies dynamics by replacing the static checkerboard placement method with a dynamic UAV flight path approach. The UAV autonomously flies through pre-set waypoints at various positions and angles around the vehicle, dynamically capturing calibration images without manual intervention. This resolves the contradiction by automating the time-consuming manual process while maintaining the requirement for multiple inclination angles.
Solution Approach 2:
The system implements self-service through automated UAV flight execution and image capture. The calibration process no longer requires manual placement of checkerboards at various inclinations; instead, the UAV autonomously navigates to waypoints and captures images, eliminating manual labor while preserving measurement precision through programmed flight paths.
2Measurement precision
If a checkerboard is placed at various inclinations throughout a visible range of a vehicle sensor for accurate calibration, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The UAV serves as an intermediary device that simplifies the calibration system. Instead of requiring complex manual operations to place and adjust checkerboards at various inclinations, the UAV autonomously positions itself and captures images. This intermediary approach reduces operational complexity while maintaining the ability to capture calibration data from multiple angles and positions.
Solution Approach 2:
The patent replaces the mechanical manual process of placing and adjusting physical checkerboards with an automated aerial system. The mechanical complexity of manual checkerboard manipulation is substituted with automated UAV navigation and image capture, reducing the complexity of human operation while preserving measurement precision.
3Measurement precision
If an indoor space including a calibration pattern is used for sensor calibration, then measurement precision is improved, but adaptability deteriorates due to spatial limitation
Solution Approach 1:
The patent transitions from two-dimensional ground-based calibration patterns to three-dimensional aerial calibration using a UAV. By moving the calibration target (checkerboard on UAV) into the air space around the vehicle, the system eliminates spatial limitations of indoor calibration rooms and enables calibration to be performed in various outdoor locations while maintaining measurement precision through controlled flight paths.
Solution Approach 2:
The UAV-based calibration system provides universality by enabling calibration to be performed in multiple locations (indoor and outdoor, various weather conditions) rather than being restricted to specific indoor calibration spaces. The same system can adapt to different environments while maintaining calibration accuracy through automated flight path execution.
4Reliability
If sensors are calibrated frequently due to dynamic vehicle characteristics and sensor fixing device physical characteristics, then reliability is improved, but loss of time increases
Solution Approach 1:
The automated UAV calibration system enables frequent recalibration without proportionally increasing time investment. The self-service nature of the system allows quick execution of calibration tasks, making it practical to perform recalibration frequently to maintain reliability as vehicle characteristics and sensor positions change over time.
Solution Approach 2:
The dynamic UAV flight path system adapts to different calibration scenarios and can be quickly re-executed when needed. This dynamic approach supports frequent recalibration to maintain sensor reliability while minimizing time loss through automated, efficient execution of calibration sequences.
Data Source
AI summary
The autonomous driving device including a communication circuit configured to communicate with an unmanned aerial vehicle, a plurality of sensors disposed in the autonomous vehicle to monitor all directions of the autonomous vehicle, and a processor, wherein the processor is configured to: control the unmanned aerial vehicle to hover at each of a plurality of waypoints of a designated flight path by controlling a relative position of the unmanned aerial vehicle through the communication circuit, change a posture angle of the unmanned aerial vehicle to a plurality of posture angles corresponding to the waypoints of the flight path, generate a plurality of images including the checkerboard and corresponding to the plurality of waypoints and the plurality of posture angles through the plurality of sensors, and calibrate the plurality of sensors on the basis of a relationship between matching points of the plurality of images.