Driving Control Apparatus Event-Based Detection and Processing Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving control methods face challenges in real-time decision-making due to the high processing load caused by a large number of traffic lines and obstacles, making it difficult to determine accurate driving actions in a timely manner.
Innovation Solution
A method that extracts relevant events based on detection information and creates a driving plan with defined actions for each event, adjusting detection conditions according to the content of the driving action, thereby reducing processing load and ensuring accurate real-time determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiscale recognition is used to calculate traffic lines of vehicle and obstacle, then the accuracy of driving action determination is improved, but the processing load increases and real-time determination becomes difficult
Solution Approach 1:
The patent segments the continuous traffic line into multiple discrete traffic line segments based on detection information. Instead of processing the entire traffic line as one continuous object, the system divides it into segments that can be processed independently and in parallel, reducing the overall processing load while maintaining accuracy in driving action determination.
Solution Approach 2:
The patent dynamically adjusts the detection condition based on the driving action being determined. The detection condition is not fixed but changes according to the specific driving scenario, allowing the system to process only the necessary information at each moment, thereby improving real-time performance without sacrificing accuracy.
2Loss of information
If uniform detection conditions are applied, then comprehensive detection information is acquired, but the processing load remains constantly high
Solution Approach 1:
The patent applies partial detection by adjusting detection conditions to acquire only the necessary amount of information for each specific driving action. Instead of uniformly detecting all possible objects and parameters in all situations, the system performs partial detection tailored to the current driving context, reducing processing load while maintaining comprehensive information acquisition where needed.
Solution Approach 2:
The patent changes detection parameters dynamically based on the driving action. The detection condition includes parameters such as detection range, detection accuracy, and detection frequency, which are adjusted according to the specific driving scenario. This allows the system to maintain comprehensive information acquisition when necessary while reducing processing load in less critical situations.
Data Source
AI summary
A driving control method is provided in which a processor configured to control driving of a vehicle acquires detection information around a vehicle on the basis of a detection condition that can be set for each point; sequentially extracts events which the vehicle encounters, on the basis of the detection information; arranges the extracted events in the order of encounters with the vehicle; creates a sequential driving plan in which a driving action is defined for each of the events on the basis of the detection information acquired in the events; executes a driving control instruction for the vehicle in accordance with the driving plan; and determines, on the basis of the content of the driving action defined for a first event which the vehicle encounters earlier, a second detection condition regarding at least one second event which the vehicle encounters after the first event.