Map Generation Curvature Estimation for Deformed Intersections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map generation technologies fail to accurately generate smooth lane shapes in deformed intersections where entry and exit points are offset, leading to potential traffic safety and convenience issues.
Innovation Solution
A map generation apparatus and system that calculates the curvature of travel trajectories using processor-based algorithms, estimates lane shapes in specific scenes like deformed intersections, and adds these shapes to the map data, even in the absence of guidance markings, by utilizing travel trajectory information from multiple vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a Bezier curve or single arc is used to connect entry and exit points of an intersection, then the lane shape can be simplified, but smooth movement cannot be achieved in deformed intersections where entry and exit are offset
Solution Approach 1:
The patent applies curvature calculation to the travel trajectory and uses this curvature information to estimate the lane shape in deformed intersections. By calculating the curvature of the actual travel path and incorporating this into the lane shape estimation, the system achieves smooth lane shapes that accurately represent the deformation, rather than forcing a simple arc or Bezier curve that cannot capture the offset between entry and exit points.
Solution Approach 2:
The patent uses travel trajectory information from actual vehicle operation as feedback to refine the lane shape estimation. By analyzing the actual paths taken by vehicles through the intersection and comparing these with the generated lane shapes, the system continuously improves the accuracy and smoothness of the lane representation, allowing the map generation apparatus to adapt to real-world deformation.
2Measurement precision
If travel trajectory information from multiple vehicles is collected and processed, then accurate lane shapes can be estimated in deformed intersections, but the system complexity and data processing requirements increase
Solution Approach 1:
The patent merges travel trajectory information from multiple vehicles by collecting trajectory data from various sources (including other vehicles' navigation systems) and processing it through a unified curvature calculation and lane shape estimation algorithm. This consolidation approach allows accurate lane shape estimation in deformed intersections by aggregating data from multiple sources while using a single coherent processing methodology.
Solution Approach 2:
The patent introduces a server device as an intermediary that collects and stores travel trajectory information from multiple vehicles before the map generation apparatus processes it. This intermediary layer simplifies the overall system architecture by centralizing data collection and storage functions, allowing the main processing system to focus on curvature calculation and lane shape estimation without handling raw data from multiple vehicles directly.
Data Source
AI summary
A map generation apparatus includes: a processor and a memory coupled to the processor. The memory is configured to store travel trajectory information indicating a travel trajectory of a vehicle. The processor is configured to perform: recognizing surrounding situation of the vehicle; calculating a curvature of the travel trajectory in a specific scene where traveling map information has not been acquired, based on the travel trajectory information stored in the memory; estimating a lane shape in the specific scene based on the curvature; and adding the lane shape to the traveling map information.