Occupancy Grid Road Branch Detection Without Lane Markings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems rely on traffic signs and lane markings, which may be absent, damaged, or incorrect, making it difficult to reliably detect locations where roads diverge or converge.
Innovation Solution
An occupancy grid-based method and apparatus that detects road boundaries to identify locations where branches of an area converge or diverge, using chains of region of interest windows to efficiently and accurately determine the number of boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lane markings and traffic signs are used to detect road branches, then detection accuracy is improved, but reliability deteriorates when markings are absent, damaged, or unclear
Solution Approach 1:
The patent introduces occupancy grids as an intermediary representation layer between raw sensor data and road branch detection. The occupancy grid converts complex visual information from lane markings and signs into simplified probability maps of free space, enabling reliable detection of road branches through geometric analysis of occupancy patterns rather than direct reliance on potentially degraded markings or signs.
Solution Approach 2:
The patent replaces the mechanical/visual system of detecting road branches through lane markings and traffic signs with a computational system based on occupancy probability analysis. Instead of relying on physical road infrastructure (markings, signs), the system uses sensor data to construct probabilistic occupancy models and detects road branches through mathematical analysis of boundary convergence and divergence in the occupancy grid.
2Reliability
If multiple sensors and complex processing are used to detect road branches without relying on markings, then reliability is improved, but device complexity increases
Solution Approach 1:
The occupancy grid system serves multiple functions simultaneously: it represents free space for navigation, enables road boundary detection, facilitates road branch identification, and provides a unified framework for integrating data from different sensor types. This multi-functionality reduces the need for separate specialized systems while maintaining high reliability in detecting road branches without relying on markings or signs.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The invention provides for a detection, in an environment of an object, of the presence of a location where road branches of an area within the environment converge or diverge. The detection is based on an occupancy grid which provides information about the probability of occupancy in the environment of the object. Boundaries of the area are detected based on values of grid cells of the occupancy grid. The number of boundaries is identified. Depending on the number of boundaries identified, the presence of a location where branches of the area converge or diverge is signalled.