Autonomous Driving Control System Using Detection Accuracy Levels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving control systems do not adequately consider detection accuracy when determining avoidance and driving courses, leading to potential hindrances in driving guidance.
Innovation Solution
An autonomous driving control system that detects the vehicle's and obstacle's positions, determines a driving strategy based on distance and computation accuracy levels, and adjusts its approach to avoid obstacles effectively, using a combination of sensors and computer algorithms to assess and respond to driving conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the driving course is determined without considering detection accuracy, then the calculation process is simple, but the appropriateness of the driving guidance decreases
Solution Approach 1:
The system changes the parameter of detection accuracy assessment by introducing computation accuracy levels (first level, second level, third level) based on distance thresholds. This allows the driving guidance system to adapt its complexity based on the reliability of detection data, resolving the contradiction between simple calculation and appropriate guidance.
Solution Approach 2:
The system dynamically adjusts the driving guidance strategy based on real-time detection accuracy. When detection accuracy is high (first computation accuracy level), the system can use simpler courses; when accuracy is low (second or third level), the system automatically selects more conservative or alternative courses, making the guidance adaptive rather than static.
2Device complexity
If the detection accuracy is not considered, then the control system is simple, but the probability of hindering driving guidance increases
Solution Approach 1:
The system introduces computation accuracy levels as a key parameter that triggers different control strategies. By comparing distance against thresholds, the system determines whether to proceed with standard avoidance courses or switch to alternative courses, thereby maintaining simple control logic while reducing hindrances through accuracy-based decision-making.
Solution Approach 2:
The control system is segmented into multiple pathways based on detection accuracy levels. Instead of a single complex control algorithm, the system divides control into simpler sub-routines: standard avoidance for high accuracy, alternative avoidance for medium accuracy, and restricted avoidance for low accuracy. This segmentation reduces overall system complexity while improving productivity.
3Speed
If the avoidance course is determined without considering computation accuracy levels, then the response time is fast, but the safety and appropriateness of avoidance maneuver decreases
Solution Approach 1:
The system performs preliminary assessment of detection accuracy by comparing distance against thresholds before determining the avoidance course. This preliminary action of categorizing detection accuracy into levels allows the system to quickly select the appropriate avoidance strategy without complex real-time calculations, maintaining fast response while ensuring appropriateness.
Solution Approach 2:
The avoidance maneuver dynamically adapts based on computation accuracy levels. The system transitions between different avoidance strategies (standard, alternative, restricted) based on real-time distance measurements, enabling fast response that is simultaneously appropriate for the current detection reliability conditions.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
According to information about an own vehicle acquired by a vehicle information acquisition unit (11) and the distance and angle of an avoidance object detected by an avoidance object detection unit (12), a peripheral information recognition unit (13) computes a position and moving amount of the avoidance object. A driving strategy determination unit (14) judges the computation accuracy of each of the position and moving amount of the avoidance object, and according to the computation accuracy of each of the position and moving amount of the avoidance object, determines a driving strategy to avoid the avoidance object. According to the determined driving strategy, autonomous driving control is carried out.