Autonomous Parking LIDAR Linear Pattern Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous parking systems face challenges in accurately determining parking spaces, especially at night due to limitations in image clarity from cameras and the need for a comprehensive method to assist vehicles in parking.
Innovation Solution
The system uses LIDAR data points to identify linear patterns corresponding to elevated parking boundary indicators, enabling the Electronic Control Unit (ECU) to autonomously park vehicles by matching these patterns with a straight-line equation and determining available parking spaces based on vehicle dimensions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cameras are used for autonomous parking, then the system can detect parking spaces, but image clarity deteriorates in low light conditions
Solution Approach 1:
The patent replaces camera-based optical detection with LIDAR-based laser ranging technology. The LIDAR unit emits laser beams and measures the time of flight of reflected light to generate precise 3D point cloud data of parking spaces, eliminating dependency on visible light conditions and ensuring reliable detection in low light or nighttime environments.
Solution Approach 2:
The system changes the detection parameter from optical intensity (camera) to time of flight (LIDAR). By measuring the time taken for laser beams to travel to and from the parking space boundaries, the system obtains accurate distance information independent of illumination conditions, thus resolving the contradiction between detection reliability and light intensity dependency.
2Measurement precision
If LIDAR data is processed using traditional algorithms, then parking space detection is achieved, but processing complexity increases
Solution Approach 1:
The patent segments the LIDAR point cloud data by identifying linear patterns that correspond to parking space boundaries. The processor divides the complex 3D data into simpler linear segments, making it easier to detect and characterize parking spaces without requiring exhaustive processing of all raw data points.
Solution Approach 2:
The system creates simplified 2D representations or models of the 3D LIDAR data that capture the essential geometric features of parking spaces. By working with these simplified copies rather than the full 3D point cloud, the processing complexity is reduced while maintaining detection accuracy.
3Extent of automation
If conventional parking assistance systems are used, then basic parking support is provided, but comprehensive autonomous parking capability is lacking
Solution Approach 1:
The patent integrates multiple functions into a single autonomous parking system: LIDAR-based environment perception, linear pattern recognition for boundary detection, processor-based parking space identification, and automated vehicle control. This multi-functional integration achieves comprehensive autonomous parking capability while managing system complexity through unified architecture.
Solution Approach 2:
The system performs autonomous parking without human intervention by automatically detecting parking spaces, calculating optimal parking positions, controlling vehicle steering and acceleration, and monitoring parking execution. The vehicle serves itself by integrating sensing, decision-making, and actuation functions within the autonomous parking system.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for accurate autonomous parking in various lighting conditions, including low light, by utilizing LIDAR data to identify and orient the vehicle within available parking spaces, enhancing the overall parking process.
Implementation Method 1
computing LIDAR data points obtained from a LIDAR unit mounted on the vehicle to collect reflections of a plurality of light rays
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to a methods and systems for autonomous parking of vehicles (103). The vehicle receives an input signal for parking the vehicle in parking premises (100) comprising a plurality of parking space (102) having an elevated parking boundary indicator (101). The vehicle (103) obtains a map of the parking premises (100). Further, the vehicle (103) receives a plurality of LIDAR data points (205) of the parking premises (100) and identifies a plurality of linear patterns from the plurality of LIDAR data points (205). Thereafter, the vehicle (103) detects at least two linear patterns (401) from the plurality of linear patterns, having a predefined length and parallelly spaced apart, indicating the elevated parking boundary indicator (101) of an available parking space, where the vehicle (103) can be autonomously parked in the available parking space.