Environmental Data Filtering for Real-Time Vehicle Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern driver assist systems face challenges in efficiently processing and transmitting environmental data due to bandwidth and computing capacity limitations, particularly in real-time decision-making for autonomous vehicle control.
Innovation Solution
A method and system employing a 'feed-backward' approach that generates and transmits only deviations from predicted environmental data, using filter criteria based on current traffic situations and object parameters, reducing the volume of data to be processed and transmitted by focusing on relevant data for decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If comprehensive environmental data are captured and transmitted in real time, then the knowledge about the vehicle environment is improved, but the bandwidth and computing capacity requirements increase beyond limits
Solution Approach 1:
The patent extracts and transmits only the essential and changed information from comprehensive environmental data. Instead of transmitting all sensor data, the system identifies and transmits only relevant objects, their parameters, and changes in traffic situations, thereby reducing data volume while maintaining decision-making quality
Solution Approach 2:
The patent applies different processing qualities to different parts of the environmental data. Critical objects and parameters receive higher processing priority and are transmitted with greater detail, while less critical data are processed with lower priority or aggregated, optimizing the balance between information completeness and data volume
2Reliability
If all sensor data are processed and transmitted, then the environmental knowledge is comprehensive, but the processing time and resource consumption increase
Solution Approach 1:
The patent performs preliminary processing of environmental data by detecting objects and determining their parameters before transmission. The system pre-identifies relevant objects and parameters that will be needed for driving decisions, filtering out unnecessary data in advance to reduce real-time processing requirements
Solution Approach 2:
The patent segments the environmental data processing into distinct stages: object detection, parameter determination, traffic situation assessment, and selective transmission. This segmentation allows the system to process different types of data with appropriate methods and priorities, reducing overall processing time while maintaining reliability
3Speed
If detailed environmental data are transmitted continuously, then the real-time decision-making capability is improved, but the transmission bandwidth requirements exceed available capacity
Solution Approach 1:
The patent implements periodic transmission of environmental data based on changes in traffic situations rather than continuous transmission. The system transmits data at intervals triggered by detected changes in object parameters or traffic conditions, reducing transmission volume while ensuring timely delivery of critical information for real-time decisions
Solution Approach 2:
The patent uses feedback mechanisms to determine what data to transmit based on the current traffic situation and decision-making needs. The system assesses the relevance of detected objects and parameters against ongoing driving tasks and transmits only the feedback-relevant data, optimizing bandwidth usage while maintaining decision-making speed
Data Source
AI summary
The invention relates to a method for providing environmental data wherein first environmental data are captured at a first time, objects being detected on the basis of the captured first environmental data, and object parameters being determined for each of the objects, wherein the object parameters are associated with the objects. In addition, a traffic situation is determined on the basis of the captured first environmental data, and filter criteria are determined according to the determined traffic situation. The filter criteria comprise prioritizations and predictions for the objects and the object parameters associated therewith. At a second time, second, updated environmental data are captured and transmission data are generated on the basis of the second, updated environmental data and the filter criteria. The transmission data are output.