Vehicle Lane Estimation Without Map Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane estimation techniques require map data, which are expensive to maintain and computationally intensive, limiting their accuracy and efficiency in determining vehicle positions relative to each other.
Innovation Solution
A method using advanced driver assistance systems (ADAS) sensors to estimate lane positions without relying on map data, processing data from sensors like RADAR, cameras, and LIDAR to determine object intersections and render a dynamic scene representation, including inter-vehicle gaps and cut-ins, using processing circuitry within vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If map data are used for lane estimation, then lane position accuracy is improved, but system cost and computational complexity increase
Solution Approach 1:
The patent extracts the essential lane estimation function from map data dependency, using only sensor data from the vehicle (cameras, LIDAR, RADAR) to determine lane positions. This removes the complex map data infrastructure while maintaining lane estimation capability through object detection and inter-vehicle gap analysis.
Solution Approach 2:
The system uses the vehicle's own sensors and detected objects (other vehicles, pedestrians) to estimate lane positions without external map data. The vehicle essentially estimates its own lane position by analyzing gaps and cut-ins relative to surrounding objects, making the system self-sufficient.
2Measurement precision
If map data are used for lane estimation, then lane position accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary object detection and classification using sensor data before lane estimation. By pre-processing sensor data to identify objects, gaps, and cut-ins, the system prepares necessary information in advance, reducing real-time computational burden while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical/computational system of map data processing with a sensor-based detection system. Instead of processing pre-stored map data, the system uses real-time sensor inputs (cameras, LIDAR, RADAR) to dynamically estimate lane positions, reducing computational overhead.
3Loss of information
If map data are used for lane estimation, then comprehensive lane information is obtained, but system cost and maintenance expenses increase
Solution Approach 1:
The system uses multi-functional sensors (cameras, LIDAR, RADAR) that serve multiple purposes: object detection, distance measurement, and lane estimation. This eliminates the need for separate map data infrastructure while achieving comprehensive lane information through a single integrated sensor system.
Solution Approach 2:
The patent replaces expensive, continuously updated map data with inexpensive, real-time sensor measurements. Instead of relying on costly map maintenance and updates, the system uses affordable sensor hardware that captures current lane conditions directly, reducing both initial cost and ongoing maintenance expenses.
Data Source
AI summary
Methods, systems, and computer readable mediums for determining a scene representation for a vehicle are provided. A disclosed method includes acquiring data from one or more sensors of the vehicle. The method further includes detecting, using processing circuitry, one or more objects in a surrounding of the vehicle based on the acquired data; determining whether an object of the detected one or more objects intersects with a vehicle lane of travel; determining whether the object is in a lane of travel of the vehicle; determining inter-vehicle gaps and cut-ins between the object and the vehicle; and rendering a scene representation including the inter-vehicle gaps and cut-ins between the object and the vehicle.