Dynamic Target Line Generation for Obstacle-Aware Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle autonomous driving systems lack a method to generate optimal dynamic target lines for different environments, which are essential for safe and efficient vehicle control during lane changes and obstacle avoidance.
Innovation Solution
A method and system that acquire obstacle information in all lanes around the vehicle, generate dynamic target lines based on lane curvature and width, and adjust vehicle control to ensure safe passage, including deceleration and lane changes when necessary, using a combination of sensors and electronic maps to determine safe driving widths and conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the vehicle maintains a fixed target line for simple control, then the control system is simple, but the vehicle cannot adapt to obstacles and environmental changes
Solution Approach 1:
The patent applies dynamics by transforming the static target line into a dynamic one that adjusts in real-time based on environmental conditions. The dynamic target line is generated by deviating from the central line of the current lane based on obstacle detection results, allowing the vehicle to adapt to changing environments while maintaining a relatively simple control framework.
Solution Approach 2:
The patent implements feedback by continuously detecting obstacles and environmental conditions, then using this information to adjust the dynamic target line. The system monitors the environment, calculates safe passage widths, and modifies the target line accordingly, creating a closed-loop control system that balances adaptability with computational efficiency.
2Measurement precision
If the vehicle constantly monitors all lanes for obstacles, then obstacle detection accuracy improves, but computational load and processing time increase
Solution Approach 1:
The patent applies segmentation by dividing the monitoring area into different ranges (first preset range and second preset range) and processing obstacles in these ranges with different levels of detail. The system first identifies obstacles in the broader first range, then focuses computational resources on the more critical second range, reducing overall processing time while maintaining detection accuracy for critical obstacles.
Solution Approach 2:
The patent implements continuous monitoring and real-time generation of dynamic target lines, ensuring that the vehicle always has an up-to-date path plan. This continuous process allows the system to maintain high detection accuracy while optimizing processing efficiency through streamlined computational steps that run continuously rather than in discrete, time-consuming batches.
3Reliability
If the vehicle generates a dynamic target line that deviates from the lane center, then obstacle avoidance capability improves, but lane discipline and traffic rule compliance may be compromised
Solution Approach 1:
The patent applies preliminary anti-action by proactively generating alternative target lines that deviate from the lane center before collisions occur. The system calculates safe passage widths and creates avoidance paths in advance, allowing the vehicle to maintain traffic discipline under normal conditions while having pre-computed emergency avoidance maneuvers ready when obstacles are detected.
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting the target line position based on the detected obstacle characteristics and safe passage width. The system modifies the target line parameters (position, curvature) only when necessary and within safe boundaries, balancing obstacle avoidance with traffic rule compliance by using parameter adjustments rather than fundamental changes to the control approach.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
Disclosed are a generation method and a generation system for a dynamic target line during automatic driving of a vehicle, and a vehicle. The generation method comprises: acquiring information of obstacles in all lanes within a pre-set range near a vehicle (S1); within a first pre-set range near the vehicle, if there is no obstacle in front of the current traveling lane of the vehicle, generating, according to the curvature and width of the current traveling lane of the vehicle, a first dynamic target line of the current traveling lane of the vehicle (S2); within the first pre-set range near the vehicle, if there is an obstacle in front of the current traveling lane of the vehicle, acquiring the safe traveling width of the current traveling lane of the vehicle (S3); and if the safe traveling width of the current traveling lane of the vehicle is greater than a pre-set safe passing width, generating, according to the safe traveling width, a second dynamic target line of the current traveling lane (S4). Optimal dynamic target lines are generated for different environments, such that traveling of the vehicle is controlled according to the generated target lines during automatic driving of the vehicle.