Convolutional Lidar Imaging Super Pixel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lidar systems face limitations in resolution for detecting road debris, long-range fast movers, and object classification, especially under degraded camera perception conditions such as precipitation or nighttime, which affects their performance in autonomous vehicles.
Innovation Solution
The implementation of a convolutional stage and post-processing stage in lidar systems to combine photodetector results using overlapping super pixels, generating high-resolution depth images and correcting edge effects through dilation and erosion operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional lidar systems are used, then the system is simple and easy to operate, but the resolution for detecting road debris, long-range fast movers, and object classification is insufficient
Solution Approach 1:
The patent divides the lidar system into multiple photodetectors arranged in a grid pattern, where each photodetector captures light signals from specific spatial regions. This segmentation allows the system to process spatial information at a finer granularity, thereby improving resolution for detecting road debris, long-range fast movers, and object classification while maintaining manageable system complexity through modular architecture
Solution Approach 2:
The patent introduces a temporal dimension by capturing multiple light signals at different times and combining them through convolutional algorithms. This temporal-spatial transformation enables the system to enhance resolution and detect features that would be invisible in single-time-point measurements, addressing the resolution limitation without proportionally increasing spatial hardware complexity
2Reliability
If camera perception is used, then object detection is possible, but performance degrades under adverse conditions such as precipitation or nighttime
Solution Approach 1:
The patent replaces camera-based optical detection with lidar-based light signal detection that operates independently of ambient light conditions. By actively emitting light signals and measuring reflected signals, the system achieves reliable detection in adverse conditions such as precipitation or nighttime, eliminating the dependency on ambient illumination that plagues camera-based systems
Solution Approach 2:
The patent changes the detection parameter from passive optical capture to active light signal measurement. By controlling the emission and detection of light signals with specific temporal and spatial characteristics, the system maintains high detection reliability across varying environmental conditions, including precipitation and nighttime operations
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 enhances the resolution of lidar systems, improving the detection of road debris and long-range objects, and maintaining effective object classification even under adverse conditions, thereby enhancing the performance of autonomous vehicles.
Implementation Method 1
the lidar system illuminates an object with light and measures the reflected light with a sensor. The reflected light is used to determine features of the object that reflected it and to determine the distance the object is from the AV
Implementation Method 2
performing operations by each of a plurality of photodetectors to facilitate measurements of an intensity of a light signal reflected off an object
Data Source
AI summary
Disclosed herein are systems, methods, and computer program products for operating a lidar system. The methods comprise: performing operations by each of a plurality of photodetectors to facilitate measurements of an intensity of a light signal reflected off an object external to the lidar system; receiving, by a processor, result values from the photodetectors that indicate measured reflected intensities of the light signal; performing, by the processor, at least one convolutional algorithm to combine different sets of the result values to produce a plurality of feature values; and generating, by the processor, at least one depth image or point cloud comprising a plurality of super pixels having values respectively set to the feature values.