Semantic Dynamic Occupancy Grids With Particle-Based Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated driving technologies lack an integrated processing of semantic information in dynamic occupancy grids, leading to inconsistent and unreliable maneuver planning.
Innovation Solution
A method to build a categorized environment model by incorporating semantic information into dynamic occupancy grids, using a set of categories and particle-based updates to ensure consistency and accuracy, enabling safer and more reliable automated driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If semantic information is integrated into dynamic occupancy grids for maneuver planning, then the reliability and safety of automated driving is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The patent merges semantic information processing directly into the dynamic occupancy grid framework by creating a unified environmental model that combines both semantic categories and occupancy probabilities. This integration eliminates the need for separate parallel processing systems, thereby improving reliability while managing complexity through unification rather than multiplication of components.
Solution Approach 2:
The environmental model is designed to serve multiple functions simultaneously: it performs traditional occupancy grid mapping, object tracking, and semantic classification all within a single unified framework. This multi-functionality allows the system to improve reliability across multiple planning aspects without proportionally increasing device complexity, as the same infrastructure supports multiple capabilities.
2Measurement precision
If semantic information is integrated into dynamic occupancy grids, then the accuracy of maneuver planning is improved, but the computational complexity increases
Solution Approach 1:
The patent segments semantic information processing into discrete categorical labels that can be efficiently stored and processed within grid cells. By dividing the semantic space into distinct categories (e.g., pedestrian, vehicle, road marking) rather than continuous descriptors, the system achieves high accuracy in environment modeling while maintaining computational efficiency through standardized, discrete data structures.
Solution Approach 2:
The system applies semantic information locally at the grid cell level, where each cell contains both occupancy probability and semantic category information. This local quality approach allows accurate representation of the environment at relevant scales without requiring global processing of all semantic data, thereby improving measurement precision while controlling computational complexity through localized operations.
3Reliability
If semantic information is integrated into dynamic occupancy grids, then the safety of automated driving is improved, but the processing time increases
Solution Approach 1:
The patent performs semantic classification and category assignment during the sensor data processing stage, before maneuver planning begins. By preliminarily organizing semantic information into categorical structures during environmental modeling, the system ensures safety-critical information is ready for immediate use in planning without requiring additional processing time during the actual maneuver decision-making process.
Solution Approach 2:
The system transforms continuous sensor data into discrete semantic categories and probability values, changing the parameter representation from continuous measurements to discrete classified states. This parameter transformation enables efficient processing and comparison during planning, as discrete categories can be quickly evaluated against planning rules and constraints, thereby improving safety while reducing processing time.
Data Source
Figure 1~4c
Figure 2
AI summary
A computer-implemented method of assisting in the operation of a vehicle is disclosed, the method comprises the steps of: with at least one sensor, sensing an environment of the vehicle thereby obtaining sensor data, and deriving spatial information of the environment and semantic information from the sensor data. Furthermore, generating a dynamic occupancy grid model, in which the sensed environment is represented as a grid consisting of a plurality of grid cells, the grid cells comprising occupying information and a dynamic state represented by a set of particles, and assigning the grid cells and the particles semantic information derived from the sensor data, wherein the semantic information is represented by a set of categories. The method further comprising the steps of predicting new particle positions on the grid; determining for the grid cells predicted semantic information based on combining the semantic information assigned to the grid cells with the semantic information assigned to the particles based on their predicted new particle positions; obtaining new sensor data, and updating the predicted semantic information assigned to the grid cells and the semantic information assigned to the particles from the new sensor data, and deriving an automated driving action based on the determined new semantic information and the dynamic state of the one or more grid cells. Also disclosed is a system for performing the computer-implemented method of assisting operation of a vehicle and the vehicle comprising said system.