Autonomous Vehicle Path Planning Using Global and Local Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving technologies face challenges in accurately controlling vehicles along planned paths, especially due to the limitations of coarse-grained perceptual information from laser sensors, which can lead to collisions with obstacles and deviations from intended routes.
Innovation Solution
The method involves generating a global path using aerial photography and a global map, followed by creating a local path based on fine-grained information from vehicle-mounted cameras, ensuring the vehicle stays on course by controlling speed and steering, and iteratively adjusting the path until the destination is reached, while ensuring constraints like orientation and distance deviations are within preset ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If coarse-grained perceptual information from laser sensors is used for path planning, then the device complexity is reduced, but the measurement precision and reliability of obstacle detection deteriorate
Solution Approach 1:
The patent divides the path planning into two segments: global path planning using laser sensor data and local path planning using camera data. This segmentation allows each sensor type to be used for its strengths while maintaining overall system simplicity.
Solution Approach 2:
The patent merges laser sensor-based global path planning with camera-based local path planning into a unified two-layer control system. This combination achieves both coarse coverage and fine-grained obstacle detection without requiring a complete overhaul of the sensor system.
2Device complexity
If only global path planning is used, then the device complexity is reduced, but the manufacturing precision of route adherence deteriorates
Solution Approach 1:
The control system is segmented into global path planning (route selection) and local path planning (precise following). This allows the system to maintain simple overall structure while achieving high precision in critical local navigation tasks.
Solution Approach 2:
The patent adds a temporal dimension to path planning by continuously updating local paths based on current vehicle position and newly captured camera images. This transforms static global planning into dynamic multi-stage planning, improving route adherence precision.
3Manufacturing precision
If fine-grained local path control is added, then the route adherence precision is improved, but the loss of time in iterative path adjustment increases
Solution Approach 1:
The system performs preliminary global path planning to establish the overall route before executing local path following. This preliminary action prevents unnecessary iterative adjustments by ensuring the global route is already optimized, reducing time loss in later adjustments.
Solution Approach 2:
The local path planning continuously adjusts the vehicle trajectory along the global path using real-time camera feedback. This continuous useful action maintains precise route adherence without requiring repeated global replanning, minimizing time loss.
4Reliability
If iterative path planning and adjustment is implemented, then the reliability of obstacle avoidance is improved, but the loss of time in multiple planning cycles increases
Solution Approach 1:
The iterative planning is segmented into local-scale iterations only, while global path planning is performed once. This segmentation maintains obstacle avoidance reliability through continuous local adjustments without the time penalty of repeated global replanning cycles.
Solution Approach 2:
The system performs partial iterative planning focused only on local path following rather than complete global replanning. This partial action achieves sufficient obstacle avoidance reliability by concentrating computational resources on critical local adjustments where obstacles are most likely to be encountered.
Data Source
AI summary
Embodiments of the present disclosure relate to a method and apparatus for controlling vehicle driving. The method includes: generating a global path of a driving site of a vehicle; executing following controlling: generating a local path of a site in front of a current position of the vehicle based on the global path, the local path following a direction of the global path, controlling the vehicle to drive along the local path until reaching an end point of the local path, determining whether the vehicle reaches an end point of the global path, and terminating the controlling if the vehicle reaches the endpoint of the global path; and continuing, in response to determining the vehicle failing to reach the end point of the global path, executing the controlling.