Vehicle LiDAR Occupancy Grid Classification for Static and Dynamic Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for classifying objects around a vehicle into static and dynamic objects require extensive training data and computational resources, limiting their efficiency in autonomous driving applications.
Innovation Solution
A method and system that utilize a dynamic occupancy grid map, generated from LiDAR data and vehicle movement information, to classify objects by determining clusters and assessing region sizes and velocity vector variances, thereby distinguishing static and dynamic objects without needing extensive training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If classification algorithms based on artificial intelligence are used to classify objects around a vehicle, then classification accuracy can be improved, but the requirement for training data and computational resources increases significantly
Solution Approach 1:
The patent extracts only the essential features needed for classification (region size, velocity vector variance) from the complex point cloud data, rather than using full AI algorithms that process all data. This selective extraction achieves classification with minimal computational resources while maintaining accuracy.
Solution Approach 2:
The patent uses simple, computationally inexpensive metrics (region size calculation, velocity vector variance) instead of expensive AI models. These simple metrics can be quickly calculated and discarded, providing an efficient alternative to resource-intensive deep learning approaches.
2Adaptability or versatility
If AI-based classification algorithms are deployed, then object classification capability is enhanced, but processing time and computational load increase
Solution Approach 1:
The patent segments the classification process into distinct steps: generating occupancy grid map, calculating velocity vectors, determining region sizes, and classifying based on these features. This segmentation allows each step to be optimized independently, improving overall processing efficiency while maintaining classification capability.
Solution Approach 2:
The patent performs preliminary calculations of velocity vectors and region sizes before the actual classification decision. This preliminary action prepares the data in advance, allowing the final classification to be made quickly based on pre-computed features, thus improving processing efficiency.
3Reliability
If comprehensive object analysis is performed to distinguish static and dynamic objects, then prediction accuracy improves, but unnecessary arithmetic operations increase
Solution Approach 1:
The patent applies different analysis depths to different objects based on their characteristics. By using velocity vector variance specifically for dynamic objects and region size for static objects, it applies the right level of analysis locally to each object type, improving prediction accuracy while avoiding unnecessary computations for objects that don't require detailed analysis.
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 effectively classifies objects with reduced computational requirements, enabling vehicles to predict dynamic object paths and avoid collisions more efficiently, while minimizing unnecessary arithmetic operations for static object prediction.
Implementation Method 1
LiDAR data received from LiDAR of the vehicle
Data Source
AI summary
An embodiment method for classifying objects around a vehicle includes generating a dynamic occupancy grid map including a plurality of cells including point data corresponding to each of a plurality of objects located around the vehicle and velocity vector information of the point data, based on LiDAR data received from a LiDAR sensor of the vehicle and information related to movement of the vehicle, determining a cluster corresponding to each of the plurality of objects on the dynamic occupancy grid map using a clustering technique, and classifying an object corresponding to the cluster into a static object or a dynamic object, based on a region size of the cluster and velocity vector information included in cells belonging to the cluster.