Vehicle Path Planning Using Risk-Based Drivable Area Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems struggle to effectively avoid obstacles while ensuring safe and smooth driving, often leading to reduced traffic efficiency and increased risk due to lane changes or stopping, especially in scenarios where lane changes are not allowed.
Innovation Solution
A control method that utilizes optimized path planning based on risk levels of obstacles and vehicles, incorporating rasterization processing and passage cost analysis to determine the lowest cost path for obstacle avoidance, reducing dependency on sensor precision and enabling safe driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle stops or follows another vehicle at low speed to avoid collision, then collision safety is improved, but traffic efficiency deteriorates
Solution Approach 1:
The system dynamically adjusts the drivable area based on real-time obstacle detection and risk assessment. When an obstacle is detected, the drivable area is dynamically expanded to include adjacent lanes or regions, allowing the vehicle to maintain motion while avoiding collision. This dynamic adaptation resolves the contradiction by enabling continuous movement (maintaining productivity) while ensuring safety through real-time boundary adjustment.
Solution Approach 2:
The patent introduces a risk level dimension to the traditional two-dimensional drivable area. By calculating risk levels for different regions and dynamically adjusting the drivable area based on these risk levels, the system creates a three-dimensional safety model (space + risk). This allows the vehicle to navigate around obstacles using lateral or longitudinal movements instead of stopping, thus maintaining traffic efficiency while ensuring collision avoidance.
2Reliability
If the vehicle performs lane change to avoid obstacle, then obstacle avoidance capability is improved, but interaction with social vehicles increases
Solution Approach 1:
The system performs preliminary risk assessment of adjacent lanes before executing lane changes. By calculating the risk level of target regions in advance and selecting lanes with lower risk levels for lane changes, the vehicle can avoid obstacles while minimizing interaction with social vehicles. This preliminary evaluation ensures that lane changes are made to safer regions, reducing the impact of uncontrollable factors.
Solution Approach 2:
The system continuously monitors the risk levels of surrounding regions and adjusts lane change decisions based on real-time feedback. By incorporating ongoing risk assessment into the lane change control logic, the vehicle can adapt to changing traffic conditions and select the safest available lane, thereby reducing interaction with social vehicles while maintaining obstacle avoidance capability.
3Measurement precision
If the vehicle relies on high sensor precision for obstacle detection, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The patent changes the parameter of risk assessment from relying solely on precise sensor measurements to incorporating multiple factors including sensor data, obstacle characteristics, and environmental context. By calculating risk levels based on a composite of these parameters rather than depending exclusively on high-precision sensing, the system achieves adequate detection accuracy with a less complex sensor configuration.
Solution Approach 2:
The system introduces risk level calculation as an intermediary layer between sensor detection and path planning. Instead of directly using high-precision sensor data to make control decisions, the patent processes this data through risk assessment to generate risk levels, which then guide the drivable area adjustment and path planning. This intermediary approach allows for adequate detection performance with reduced sensor complexity by focusing computational resources on risk evaluation rather than ultra-precise measurement.
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
A control method, a related device, and a computer-readable storage medium are disclosed. The method includes: first obtaining vehicle information, obstacle information, and a drivable area of a vehicle (S301); and performing first processing based on the vehicle information and the obstacle information with reference to the drivable area of the vehicle (S302), where the vehicle information, the obstacle information, and the drivable area of the vehicle indicate M planned paths, and M is an integer greater than 0; and the M planned paths correspond to respective passage costs, and the passage costs are related to at least one of the following information: a risk level of an obstacle and a risk level of the vehicle, where the risk level of the obstacle represents a possible intrusion of the obstacle into the drivable area of the vehicle and a degree of a possible loss caused by the obstacle to the vehicle, and the risk level of the vehicle represents a degree of a possible loss caused by the vehicle to the obstacle. In the foregoing method, the obstacle can be effectively avoided, thereby ensuring safe and smooth driving of the vehicle.