Vehicle Collision Threat Assessment Using Fused Occupancy Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current dynamic movement models for vehicle collision threat assessment in complex traffic scenarios are inaccurate at longer time horizons due to the inability to integrate all context information, leading to challenges in predicting future maneuvers of multiple road users and interactions in urban driving environments.
Innovation Solution
A computer-implemented method that obtains and fuses context information to determine ego and road user occupancy information across multiple future locations and time points, using a grid-map representation to estimate collision threats, and triggers ADAS functionalities based on predetermined thresholds, potentially reducing unnecessary warnings and improving safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dynamic movement model is used to predict future trajectories, then short-term predictions are relatively accurate, but prediction accuracy deteriorates significantly at longer time horizons
Solution Approach 1:
The prediction task is segmented into two independent occupancy prediction streams: one for the ego vehicle and one for other road users. Each stream processes context information separately to generate occupancy probabilities for future locations, which are then combined to assess collision threat. This segmentation allows each predictor to focus on specific behavioral patterns while integrating comprehensive context information.
Solution Approach 2:
The patent transitions from predicting continuous trajectories to predicting discrete occupancy grids at multiple future time points. Instead of forecasting continuous path coordinates, the system evaluates probability of occupancy for each grid cell at each future time step, adding a probabilistic dimension to the prediction and enabling better handling of uncertainty in long-term predictions.
2Adaptability or versatility
If context information is not integrated, then the prediction model is simpler, but the ability to handle complex traffic scenes and interactions deteriorates
Solution Approach 1:
The occupancy prediction framework serves multiple functions simultaneously: it predicts future positions of the ego vehicle, predicts future positions of other road users, and enables collision threat assessment. By using a unified occupancy grid representation, the system handles diverse traffic scenarios including intersections, roundabouts, and multi-lane roads without requiring scenario-specific models.
Solution Approach 2:
The occupancy grid acts as an intermediary representation between raw context information and collision threat assessment. Context information about road geometry, traffic signs, and other road users is integrated into the occupancy prediction, which then serves as the basis for evaluating collision risk. This intermediary layer simplifies the overall system architecture while capturing complex interactions.
Data Source
AI summary
A computer-implemented method for collision threat assessment of a vehicle includes obtaining context information for the surrounding of the vehicle, including information about at least one road user. The method includes determining ego occupancy information for multiple possible future locations of the vehicle at multiple future points in time based on the context information. The method includes determining road user occupancy information for multiple possible future locations of the at least one road user at multiple future points in time based on the context information. The method includes fusing the ego occupancy information and the road user occupancy information to obtain fused occupancy information at each future point in time. The method includes determining a collision threat value based on the fused occupancy information.