Sensor Data Preselection for Autonomous Vehicle Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicles face limitations in public traffic areas due to regulatory requirements that demand driver intervention, and existing sensors have limited capability for efficient surroundings detection, hindering fully autonomous operation.
Innovation Solution
A method that utilizes additional sensors like LIDAR, radar, and camera sensors to preselect relevant image details and position information, transforming them into a common coordinate system for rapid and efficient object recognition, reducing processing load and enhancing reliability through coordinated sensor data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional sensors are used for surroundings detection, then the system is simpler, but the evaluation speed and detection capability are limited
Solution Approach 1:
The patent segments the sensor system into multiple specialized sensors (LIDAR for 3D mapping, radar for distant object detection, camera for detailed imaging) and divides the detection task into separate processing stages. Each sensor type handles specific detection functions, allowing parallel processing and improving overall evaluation speed without requiring a single overly complex sensor system.
Solution Approach 2:
The patent merges data from multiple sensor types (LIDAR, radar, camera) into a unified detection system. By combining the complementary strengths of each sensor and integrating their data streams, the system achieves faster and more comprehensive surroundings evaluation than any single sensor could provide alone.
2Measurement precision
If all sensor data is processed, then detection completeness is improved, but processing load and time increase
Solution Approach 1:
The patent performs preliminary actions by pre-processing sensor data to identify and filter out relevant information before final object recognition. LIDAR data is pre-processed to create 3D maps and identify potential objects, radar data is pre-processed to detect distant objects, and camera data is pre-processed to identify objects of interest. This preliminary sorting and filtering reduces the amount of data that needs detailed processing, thereby reducing overall processing time while maintaining detection completeness.
Solution Approach 2:
The patent applies local quality by processing different sensor data with different levels of detail based on their specific characteristics and detection needs. LIDAR data receives processing focused on 3D spatial information, radar data on distance and velocity, and camera data on detailed imaging. This targeted processing approach ensures accurate object recognition without uniformly processing all data at maximum detail, thus reducing processing time.
3Reliability
If multiple sensors are integrated, then detection reliability is improved, but system complexity and data transmission load increase
Solution Approach 1:
The patent implements multi-functionality by designing a unified processing system that handles data from multiple sensor types (LIDAR, radar, camera) through common processing algorithms and coordinate transformation frameworks. The system uses a universal coordinate system that can accommodate data from all sensor types, and a unified object recognition module that processes data from any sensor. This approach improves reliability through multi-sensor fusion while managing complexity through standardized processing procedures.
Solution Approach 2:
The patent introduces intermediary processing layers that mediate between raw sensor data and final object recognition. Coordinate transformation modules serve as intermediaries that standardize data from different sensor coordinate systems into a common reference frame. Data fusion algorithms act as intermediaries that integrate information from multiple sensors. These intermediary layers simplify the complexity of direct multi-sensor integration and reduce data transmission requirements by processing and filtering data at intermediate stages.
Data Source
AI summary
A method for selecting a detail for a surroundings detection by a sensor based on sensor data. The surroundings are detected by at least one additional sensor, and an object recognition is carried out based on the ascertained sensor data of the at least one additional sensor. Pieces of position information from at least one recognized object are transformed into a coordinate system of the sensor, based on the object recognition, and based on the transformed pieces of position information, the sensor uses a detail of a scanning area for the surroundings detection, or an image detail from already detected sensor data, for an evaluation. A control device is also described.