Automated Ego Vehicle Behavior Planning With Partial Situation Hierarchy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for behavior planning in automated vehicles often result in suboptimal decisions due to overly restrictive boundary conditions, leading to inefficient computing and potential safety issues.
Innovation Solution
A method that decomposes traffic situations into partial situations using a specialization hierarchy, instantiating only necessary partial situations and utilizing a tree or directed acyclic graph structure to optimize evaluation, reducing unnecessary computations and ensuring safety guarantees.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all identified partial situations are instantiated and evaluated, then safety guarantees are maintained, but computing time increases and efficiency decreases
Solution Approach 1:
The patent segments the evaluation process by introducing a specialization hierarchy that divides partial situations into different levels. Only partial situations at the current evaluation level and below are instantiated, while higher-level partial situations are skipped. This segmentation allows the system to maintain safety guarantees for relevant situations while avoiding unnecessary computation on already-addressed higher-level situations, thus reducing computing time without compromising reliability.
Solution Approach 2:
The patent applies partial action by evaluating only the necessary portion of partial situations based on the specialization hierarchy. Instead of instantiating all identified partial situations (excessive action), the system selectively instantiates only those at the current level and below (partial action). This approach maintains sufficient safety guarantees while avoiding the computational overhead of evaluating all possible partial situations, thereby improving efficiency.
2Productivity
If a specialization hierarchy is introduced to reduce instantiations, then computing efficiency improves, but system complexity increases
Solution Approach 1:
The specialization hierarchy is pre-computed and stored before the actual behavior planning process. This preliminary action organizes the partial situations into a hierarchical structure with defined levels and relationships, allowing the system to efficiently navigate and select which partial situations to instantiate during runtime. By preparing this structure in advance, the system reduces runtime complexity while maintaining high computing efficiency during actual evaluation.
Solution Approach 2:
The system dynamically adjusts the evaluation process based on the specialization hierarchy by determining a current level and selectively instantiating partial situations at that level and below. This dynamic approach allows the system to adapt its computation depth to the specific situation, optimizing the balance between efficiency and complexity management rather than using a fixed evaluation strategy for all cases.
Data Source
AI summary
A computer-implemented method for planning the behavior of an at least partially automated EGO vehicle. The method uses a data basis of predefined partial situations and an evaluation model for each of these partial situations as well as a predefined set of rules for assessing possible behaviors of the EGO vehicle in a given situation. The method includes decomposing the given situation into a set of partial situations of the data basis and determining boundary conditions for the behavior of the EGO vehicle on the basis of the evaluation models of the identified partial situations. In doing so, a specified specialization hierarchy for the partial situations of the data basis is used.