Autonomous Lane Change Path Planning Using Camera and Map Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle path planning systems for semi-autonomous or autonomous vehicles are limited by the effective viewing distance of forward-looking cameras, which restricts their ability to accurately perform lane changing maneuvers at higher speeds due to insufficient detection of roadway markings and curvature, leading to potential harshness and inaccuracies in path generation.
Innovation Solution
The system employs a method that separates the roadway into segments, using camera measurements for the first segment and map database values for segments beyond the camera's range, ensuring a smooth transition and extending the effective range for path planning, allowing for more accurate lane centering and changing by determining the desired path using fifth-order polynomial equations and continuity constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera measurements are used for path planning, then measurement precision is improved within camera range, but the effective viewing distance is limited beyond 80 meters
Solution Approach 1:
The patent combines camera measurements for near-field path planning (first segment) with map database roadway points for far-field path planning (second segment). This merging allows the system to overcome the camera's limited 80-meter viewing distance by integrating complementary data sources, enabling accurate path generation at highway speeds while maintaining measurement precision within the camera's effective range.
2Measurement precision
If camera range is extended to detect roadway curvature, then path planning accuracy is improved, but device complexity increases
Solution Approach 1:
The patent uses map database roadway points as an intermediary to extend the effective detection range for roadway curvature. Instead of relying solely on the camera to detect distant roadway features, the system accesses pre-stored map data to provide curvature information beyond the camera's 80-meter limit, thereby improving path planning accuracy without significantly increasing sensor system complexity.
3Productivity
If lane changing is performed at higher speeds, then productivity is improved, but path generation accuracy deteriorates due to sensor latency
Solution Approach 1:
The patent performs preliminary path planning by dividing the roadway into segments and pre-calculating desired paths using both camera measurements and map database points. This allows the system to generate accurate predicted paths in advance, compensating for sensor latency during actual lane changing maneuvers at highway speeds, thereby maintaining path generation accuracy while improving productivity.
Data Source
AI summary
A system and method for providing path planning and generation for automated lane centering and/or lane changing purposes for a vehicle traveling on a roadway, where the method employs roadway measurement values from a vision camera within an effective range of the camera and roadway measurement values from a map database beyond the range of the camera. The method uses the roadway measurement values from the camera to determine a desired path along a first segment of the roadway and identifies an end of the first segment based on how accurately the camera defines the roadway. The method then uses the roadway measurement values from the map database to determine the desired path along a second segment of the roadway that begins at the end of the first segment, where a transition from the first segment to the second segment is smooth.