Dynamic Sensor Recognition Area Adjustment for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous driving systems are limited by fixed recognition ranges and hardware performance of sensors, which restrict their ability to adapt to varying driving conditions and situations.
Innovation Solution
A vehicle system that includes a communication part, driving part, and information acquisition part with a control part that dynamically adjusts the recognition areas and performance of sensors like radar, LiDAR, and cameras based on road conditions, travel information, and recognition results to optimize sensor performance for specific tasks such as improving accuracy, acquiring information about moving objects, or enhancing classification characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed recognition range and fixed hardware performance of sensors are used, then the system structure is simple and reliable, but the recognition performance is limited and cannot adapt to varying driving conditions
Solution Approach 1:
The patent implements dynamic adjustment of sensor recognition ranges based on driving conditions. The control part receives recognition results from sensors and dynamically changes recognition areas of the radar and LiDAR according to the required performance, transforming fixed sensor operation into adaptive dynamic operation that responds to real-time driving scenarios
Solution Approach 2:
The patent changes operational parameters of sensors (recognition range, recognition area) based on driving conditions and recognition results. The control part adjusts these parameters dynamically to optimize recognition performance for different situations, such as changing radar recognition area to vicinity of moving objects or LiDAR recognition area to center of specific areas for improved resolution
2Measurement precision
If the recognition area of sensors is expanded to cover all surrounding areas, then the recognition coverage is improved, but the recognition accuracy for specific areas is reduced
Solution Approach 1:
The patent applies different recognition qualities to different spatial regions. Instead of uniform recognition across all areas, the system dynamically adjusts recognition areas to concentrate high-quality recognition resources on specific regions of interest (such as vicinity of moving objects or center of target areas), while maintaining broader but lower-priority coverage elsewhere
Solution Approach 2:
The recognition area configuration is dynamically adjusted based on real-time driving conditions and recognition results. The control part can shift recognition focus from broad coverage to concentrated high-accuracy monitoring of specific areas as situations evolve, optimizing the trade-off between coverage and accuracy continuously
3Measurement precision
If multiple sensors operate at maximum performance simultaneously, then the recognition accuracy is maximized, but the energy consumption and system complexity increase
Solution Approach 1:
The patent implements partial action by activating or adjusting sensor performance levels based on actual needs. Instead of all sensors operating at maximum capacity continuously, the control part selectively adjusts recognition areas and performance levels of individual sensors (radar, LiDAR, camera) according to specific driving conditions and recognition requirements, reducing unnecessary energy consumption while maintaining adequate recognition accuracy
Solution Approach 2:
The system dynamically changes operational parameters of sensors including recognition range, recognition area, and performance levels. Based on driving conditions and recognition results, the control part adjusts these parameters to optimize the balance between recognition accuracy and energy consumption, scaling sensor performance up or down as situations require rather than maintaining constant maximum operation
Data Source
AI summary
Provided is a vehicle capable of performing safe autonomous driving by selecting a recognition area of a sensor for performing autonomous driving and maximizing the performance of the sensor according to a situation. The vehicle for performing autonomous driving includes a communication part, a driving part configured to drive the vehicle and acquire information about an element that drives the vehicle, an information acquisition part including a camera, a radar and a LiDAR, and a control part. The control part is configured to determine road condition information of a road on which the vehicle travels based on a signal acquired from the communication part, determine travel information of the vehicle based on information acquired from the driving part, receive a recognition result of the information acquisition part, determine a required performance based on the road condition information, the vehicle travelling information, and the recognition result, and change an object recognition.