Vehicle Control Command Checks Using Safe Sets for Future Maneuvers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for vehicle control fail to ensure that situation avoidance maneuvers (SAMs) can be executed safely, particularly when key sensors fail, leading to potential unsafe vehicle states due to unaccounted disturbances.
Innovation Solution
A method to determine if a vehicle control command precludes a future SAM by obtaining safe sets representing successful maneuver states, predicting future vehicle states, and comparing them to these safe sets, adjusting for disturbances and vehicle dynamics, allowing for off-line calculations to optimize real-time decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system executes situation avoidance maneuvers immediately when sensors fail, then vehicle safety is improved, but transport efficiency deteriorates due to unnecessary maneuvers
Solution Approach 1:
The system performs preliminary evaluation by predicting future vehicle states and comparing them against pre-computed safe sets before executing maneuvers. This preliminary action distinguishes between situations that truly require immediate intervention and those where normal operation can continue, avoiding unnecessary maneuvers that would disrupt transport efficiency while maintaining safety for genuinely critical situations
Solution Approach 2:
The system continuously monitors vehicle state and compares predicted future states against safe sets, creating a feedback mechanism that dynamically adjusts maneuver execution decisions. This feedback loop ensures maneuvers are executed only when necessary while maintaining transport efficiency during normal operation
2Measurement precision
If the system performs real-time computational analysis to determine safe maneuver execution, then decision accuracy is improved, but computational load increases
Solution Approach 1:
The system performs computationally intensive safe set calculations offline before real-time operation. During real-time decision-making, the system only needs to compare current and predicted states against these pre-computed safe sets, dramatically reducing online computational load while maintaining high decision accuracy through rigorous mathematical validation
Solution Approach 2:
The computational task is segmented into offline preparation (computing safe sets using detailed vehicle models and disturbance scenarios) and online execution (comparing states against safe sets). This segmentation moves heavy computation to when it's not time-critical, enabling accurate real-time decisions with minimal computational burden
3Reliability
If the system accounts for disturbances and vehicle dynamics in safety determination, then safety reliability is improved, but device complexity increases
Solution Approach 1:
The system pre-computes safe sets that inherently account for disturbances and vehicle dynamics using detailed mathematical models. By performing this complex analysis beforehand, the real-time system achieves high safety reliability through rigorous disturbance consideration without requiring complex real-time computation or additional sensors
Solution Approach 2:
The system transforms the complex problem of real-time disturbance compensation into a simpler state comparison problem. By changing parameters from continuous disturbance modeling to discrete safe set membership testing, the system maintains high safety reliability while reducing real-time computational complexity
Data Source
Figure 1~2
Figure 3~5
Figure 6~7
AI summary
A method for determining if a vehicle control command for controlling a vehicle (1) associated with a current vehicle state precludes a future situation avoidance maneuver (SAM) by the vehicle.The method comprises obtaining one or more safe sets, wherein each safe set represents a range of vehicle states from which a future SAM can be initialized with prospect for success. The method also comprises obtaining the current vehicle state and the control command and predicting a future vehicle state based on the current vehicle state and on the control command. The method comprises comparing the predicted future vehicle state to the one or more safe sets and determining that the control command precludes the future SAM if the predicted future vehicle state is not comprised in any of the one or more safe sets.