Lidar Grid Velocity Estimation Using Image-Based Feature Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle navigation systems inadequately predict the behavior of detected objects, leading to insufficient projection accuracy for navigation, which hinders the development of reliable autonomous driving solutions.
Innovation Solution
The implementation of a method that converts lidar grids into images and applies image processing techniques to estimate object velocities, reducing processing time and bandwidth, thereby enhancing the accuracy and speed of collision probability assessments and enabling more efficient navigation decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional autonomous vehicle navigation systems are used, then the system can operate with existing technology, but the projection accuracy of detected objects is insufficient
Solution Approach 1:
The patent replaces conventional lidar grid processing with image processing techniques. Lidar grids are converted into image format, allowing the application of成熟的image processing algorithms to estimate object velocities more accurately. This substitution improves projection accuracy by leveraging specialized image analysis methods rather than general-purpose lidar processing.
Solution Approach 2:
The patent transforms the data representation from lidar grid format to image format, changing the parameter structure to enable velocity estimation through image processing. This parameter transformation allows extraction of motion information that was not readily available in the original lidar grid representation, thereby improving projection accuracy.
2Measurement precision
If detailed lidar grid processing is performed to improve velocity estimation, then measurement precision improves, but processing time and bandwidth increase
Solution Approach 1:
The patent substitutes complex lidar grid analysis with more efficient image processing operations. By converting lidar data to image format, the system can use optimized image processing hardware and algorithms that require less computational time and bandwidth while maintaining or improving velocity estimation accuracy.
Solution Approach 2:
The patent creates an image copy of the lidar grid data, processing this copy rather than the original lidar grid. This copying approach allows the system to apply specialized image processing techniques that are computationally more efficient, reducing processing time and bandwidth requirements while achieving the same velocity estimation goals.
3Measurement precision
If conventional navigation systems are used, then system complexity is lower, but collision probability assessment accuracy is insufficient
Solution Approach 1:
The patent replaces conventional collision assessment methods with image processing-based velocity estimation. This substitution improves collision probability assessment accuracy by providing more reliable velocity data, while the use of standard image processing techniques keeps the added complexity manageable.
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 improves the accuracy of object trajectory projections, reduces power consumption, and enhances navigation safety by providing timely and accurate collision probability assessments, allowing autonomous vehicles to make faster and more informed decisions.
Implementation Method 1
Individual points are measured by generating a laser pulse and detecting a returning pulse, if any, reflected from an environmental object, and determining the distance to the reflective object according to the time delay between the emitted pulse and the reception of the reflected pulse
Implementation Method 2
One such sensor is a light detection and ranging (lidar) device. A lidar device actively estimates distances to environmental features while scanning through a scene
Data Source
AI summary
Systems and methods are described for measuring velocity of an object detected by a light detection and ranging (lidar) system. According to some aspects a method may include receiving a lidar dataset generated by the lidar system, transforming the lidar dataset into a first layer dataset and a second layer dataset, and converting the first layer dataset into a first image and the second layer dataset into a second image. The method may also include performing a feature detection operation that identifies at least one feature in the first image and the same feature in the second image, locating a first location of the feature in the first image and a second location of the feature in the second image, and generating a velocity estimate of the feature based on a difference between the first location and the second location and a difference between the different time intervals.