Sensor Fusion Processing for Weak-Reflection Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated driving and advanced driver assistance systems face challenges in accurately recognizing objects in various environments due to the limitations of individual sensors, such as cameras and millimeter wave radars, which deteriorate in adverse weather conditions or at night, leading to incomplete recognition of objects with weak reflection intensity.
Innovation Solution
An information processing device that performs fusion processing on data from multiple sensors, including cameras and millimeter wave radars, to enhance recognition accuracy by combining detection signals, distance, speed, and angle data before and after recognition, and adaptively using early or late fusion techniques based on environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single sensor type is used for object detection, then the device complexity is reduced, but the recognition accuracy deteriorates in various environmental conditions
Solution Approach 1:
The patent combines multiple sensor types (camera, millimeter wave radar, laser radar) into a unified sensor system. Each sensor type detects objects using different physical principles and has different strengths, and by merging their detection results through fusion processing, the system achieves reliable object recognition across various environmental conditions while maintaining manageable system complexity through integrated architecture
2Measurement precision
If multiple sensors are combined for fusion processing, then the object recognition accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent segments the sensor fusion process into distinct functional modules: individual sensor detection units, feature extraction modules, data association modules, and fusion processing units. This segmentation allows each component to be optimized independently and simplifies the overall system architecture by breaking down the complex fusion task into manageable stages, thereby reducing system complexity while maintaining high recognition accuracy
Solution Approach 2:
The patent introduces intermediary processing layers between raw sensor data and final object recognition results. These intermediaries include detection result generation modules that create standardized output formats from different sensor types, and data association modules that serve as mediators to correlate detections across sensors before fusion, thereby managing complexity by creating structured intermediate representations
3Reliability
If fusion processing is performed on raw detection signals, then the object recognition reliability is improved in adverse conditions, but the processing time increases
Solution Approach 1:
The patent performs preliminary processing of sensor data including detection result generation and feature extraction before the actual fusion processing. By preparing standardized detection results and extracting key features in advance from each sensor type, the system reduces the computational burden during time-critical fusion operations, thereby maintaining high recognition reliability in adverse conditions while minimizing processing time
Solution Approach 2:
The patent implements dynamic fusion processing that adapts to environmental conditions and detection scenarios. The system dynamically adjusts the extent and method of fusion processing based on factors such as weather conditions, object type, and detection confidence levels, allowing it to perform comprehensive fusion when reliability is most needed while using simplified processing when conditions are favorable, thus optimizing the balance between reliability and processing time
Data Source
AI summary
There is provided an information processing device that processes detection information of an external recognition sensor.The information processing device includes: a recognition unit that performs recognition processing on an object on the basis of a detection signal of the sensor; and a processing unit that performs fusion processing on first data before the recognition by the recognition unit and another data. The information processing device further includes a second recognition unit that performs recognition processing on the object on the basis of a detection signal of a second sensor. The processing unit performs fusion processing on third data before the recognition by the second recognition unit and the first data, fusion processing on fourth data after the recognition by the second recognition unit and the first data, and the like.