Autonomous Vehicle Trajectory Selection With Hierarchical Rule Priorities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles face challenges in executing complex decisions due to the difficulty in navigating diverse trajectories that satisfy a hierarchical plurality of rules, leading to suboptimal short-term and long-term driving behaviors.
Innovation Solution
A signal processing system selects trajectories for autonomous vehicles using a hierarchical plurality of rules, maintaining diverse trajectories and prioritizing rule violations based on their priority levels to generate paths that satisfy these rules, thereby improving navigation efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the system maintains diverse trajectories to explore multiple possible paths, then the adaptability and decision-making quality improve, but the computational complexity and processing time increase
Solution Approach 1:
The trajectory generation is divided into multiple planning steps, where each step generates a subset of diverse trajectories. The system segments the complex decision space into manageable portions by maintaining only k diverse trajectories at each step rather than evaluating all possible trajectories, thus reducing computational complexity while preserving adaptability.
Solution Approach 2:
The system dynamically adjusts the diversity of trajectories maintained at each planning step based on the current situation. By using a parameter k to control the number of diverse trajectories and allowing the set of diverse trajectories to evolve across planning steps, the system adapts its computational effort to the complexity of the decision-making situation.
2Reliability
If the system evaluates all possible trajectories to ensure rule compliance, then the reliability of rule satisfaction improves, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary evaluation of trajectories at each planning step by checking rule violations in advance. By identifying and pruning trajectories that violate high-priority rules early in the planning process, the system avoids wasting computational resources on evaluating complete paths that would ultimately be rejected, thus reducing processing time while maintaining reliability.
Solution Approach 2:
The system uses feedback from rule violation detection to guide trajectory generation. When a trajectory violates a rule, the system adjusts subsequent trajectory generation to avoid similar violations, using the hierarchical priority information to focus computational effort on trajectories that are more likely to satisfy all rules, thereby improving efficiency without sacrificing reliability.
3Reliability
If the system prioritizes avoiding rule violations in early planning steps, then the overall path compliance improves, but the flexibility to find optimal paths decreases
Solution Approach 1:
The system applies different evaluation criteria at different planning steps by using hierarchical priority levels for rules. High-priority rules are enforced strictly in early planning steps to ensure fundamental compliance, while lower-priority rules allow more flexibility in later steps for optimization, thus maintaining both compliance and adaptability through localized quality control at each decision level.
Solution Approach 2:
The system dynamically balances compliance and optimization by adjusting the strictness of rule enforcement across planning steps. In earlier steps, the system maintains diverse trajectories with relaxed enforcement to preserve flexibility, then progressively applies stricter compliance checks as planning progresses, allowing the system to adapt its behavior to the stage of planning and find optimal compliant paths.
Data Source
AI summary
Provided are methods for trajectory generation based on a hierarchical plurality of rules using diverse trajectories, which can include generating a first set of trajectories for a vehicle from a first pose, identifying a first trajectory and a second trajectory from the first set of trajectories, generating a second set of trajectories for the vehicle from a second pose and a third set of trajectories for the vehicle from a third pose, identifying a third trajectory based at least in part on the second set of trajectories and the third set of trajectories, the third trajectory violating a first behavioral rule associated with a first priority that is less than a priority of behavioral rules violated by other trajectories, and determining a path for the vehicle based at least in part on the third trajectory. Systems and computer program products are also provided.