Autonomous Vehicle Trajectory Hierarchy for Safety Stop Release
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in safely transitioning back to nominal operating mode after encountering faults, such as sensor failures, due to the risk of unsafe trajectory selection by AI systems, necessitating a robust mechanism for validating and prioritizing trajectories to ensure collision avoidance and passenger safety.
Innovation Solution
The implementation of a trajectory manager process (TMP) that evaluates trajectories based on punctuality, consistency, feasibility, and collision checking criteria, prioritizing safety by selecting valid and collision-free paths and employing a hierarchy of trajectories, including emergency stops, to ensure the vehicle's safe operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the AI system selects trajectories autonomously, then the responsiveness and automation of the vehicle are improved, but the reliability and safety may deteriorate due to potential unsafe trajectory selections
Solution Approach 1:
The patent introduces a trajectory manager process as an intermediary between the AI system and the vehicle control systems. This intermediary validates AI-generated trajectories against safety criteria (collision detection, feasibility constraints, consistency checks) before execution, and can override unsafe selections. It also manages a hierarchy of backup trajectories (contingent, stored, transformed) that activate when AI trajectories fail validation, thus maintaining automation while ensuring reliability through layered validation and fallback mechanisms.
2Reliability
If the system implements comprehensive trajectory validation and hierarchy management, then the safety and reliability are improved, but the device complexity increases
Solution Approach 1:
The trajectory management system is segmented into distinct functional modules: AI trajectory generation, validation subsystem (checking punctuality, consistency, feasibility), collision detection, hierarchy management (contingent, stored, transformed trajectories), and override mechanisms. This modular segmentation allows comprehensive safety validation while organizing complexity into manageable, independent components that can be developed and tested separately.
Solution Approach 2:
The system performs preliminary validation of AI-generated trajectories before they are executed, checking multiple safety criteria in advance. It also pre-generates and stores contingent and stored trajectories as backup options, and pre-establishes transformation rules for modifying trajectories. These preliminary actions ensure safety is verified before critical decisions are made, reducing the complexity of real-time emergency responses.
3Reliability
If the vehicle requires manual release signals after safety stops, then the reliability is improved by preventing premature resumption, but the productivity and response time decrease
Solution Approach 1:
The system implements feedback mechanisms where the trajectory manager continuously monitors system state and provides status information to remote operators. When a safety stop is triggered, the system waits for explicit release signals from remote operators based on feedback about the current situation. This feedback loop ensures that resumption decisions are informed by real-time system state assessment, preventing premature mode switching while allowing efficient recovery when conditions permit.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and apparatus for selecting among a plurality of trajectories comprises receiving at least a first and second trajectory, each having an associated level in a trajectory hierarchy, determining whether the first trajectory is viable, determining a present level limit for trajectory selection, and if the first trajectory is viable and its level does not exceed the present level limit, executing the first trajectory. If the first trajectory is not viable or the associated level of the first trajectory exceeds the present level limit, the second trajectory is executed if it is viable and its associated level does not exceed the present level limit. A state variable is set such that when a trajectory lower in a trajectory hierarchy is executed, a trajectory selector waits for a message from a monitor or an operator before returning to the use of higher level trajectories.