Vehicle Trajectory Determination Using Occupancy Map Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle trajectory determination systems do not adequately account for the behavior of other road users, leading to suboptimal collision avoidance and traffic risk reduction.
Innovation Solution
A method for determining and monitoring a target vehicle trajectory that considers all drivable trajectories, including those of other road users, by using an occupancy map and state space analysis to identify collision-free paths and adjust the trajectory based on changes in road user plans and environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle trajectory determination system only considers the vehicle's own drivable trajectories, then the system complexity remains lower, but the collision avoidance capability and traffic risk reduction are suboptimal
Solution Approach 1:
The system segments the trajectory determination problem into two parts: first determining the vehicle's own drivable trajectories based on its state and environment, then separately determining the trajectories of other road users. This segmentation allows the system to incorporate other road users' behavior without creating a single monolithic complex system, as each trajectory determination can be performed independently and then integrated through the occupancy map.
Solution Approach 2:
The occupancy map serves as an intermediary data structure that integrates information about the vehicle's drivable trajectories and other road users' trajectories. Instead of directly complex interactions between multiple trajectory planning systems, the occupancy map mediates by representing occupied and free spaces, allowing the vehicle's trajectory determination to account for other road users in a structured way that manages overall system complexity.
2Reliability
If the system incorporates trajectories of other road users to improve collision avoidance, then the traffic risk reduction improves, but the computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary determination of other road users' trajectories before finalizing the vehicle's target trajectory. By first identifying the drivable trajectories of other road users and updating the occupancy map with their intended paths, the system prepares the environmental context in advance. This preliminary action allows the vehicle's trajectory determination to proceed more efficiently by working with pre-processed occupancy information rather than calculating everything simultaneously.
Solution Approach 2:
The occupancy map is dynamically updated based on the determined trajectories of other road users and the vehicle's current state. Rather than using static environmental models, the system continuously adapts the occupancy map to reflect current traffic conditions and predicted movements. This dynamic approach allows the trajectory determination to account for changing conditions without requiring complete recalculation from scratch, reducing processing time while maintaining accuracy.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
An automatic determination and/or monitoring of a target trajectory (8) for a vehicle (10), which connects a start point (SP) corresponding to the current position of the vehicle (10) with a destination point (sTP), comprises: Determining various trajectories (1, 7) of the vehicle (10) that connect the start point (SP) with the destination point (sTP). Identifying one additional target trajectory each for road users (21, 22), each of which connects the start point of the respective road user (21) with a destination point corresponding to the respective road user (21; 22). Determining those of the vehicle's (10) trajectories as collision-free trajectories (1) that will not result in a collision with any of the other road users (21, 22) when the respective road user (21; 22) is moving along its target trajectory.Determining and/or monitoring the target trajectory of the vehicle (1) depending on the collision-free trajectories (1) of the vehicle (10).