Steering Actuator Capability Estimation for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems fail to accurately estimate and adjust steering actuator capabilities in real-time based on dynamic vehicle and environmental conditions, leading to potential collisions or maneuverability issues due to over- or under-predicted path planning.
Innovation Solution
A system that estimates rack force and motor health in real-time using a combination of a lookup table and a trained model, such as a neural network, to calculate maximum achievable steering wheel angle, velocity, and acceleration, and adjusts the planned path accordingly to align with the vehicle's capabilities, considering factors like temperature, tire pressure, and surface friction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If path planning is over-predicted to improve maneuverability, then the vehicle can execute more aggressive paths, but the actual path may cause collision with obstacles
Solution Approach 1:
The system continuously monitors actual steering actuator performance against predicted capabilities and adjusts path planning in real-time. Sensors provide feedback on actual steering angle, velocity, and acceleration, which are compared with predicted values to dynamically adjust the path plan and prevent collisions while maintaining maneuverability.
Solution Approach 2:
The path planning system transitions from static predicted capabilities to dynamic real-time capability estimation. The system continuously updates the achievable angle, velocity, and acceleration based on current vehicle state and environmental conditions, allowing the path plan to adapt dynamically rather than relying on fixed predictions.
2Reliability
If path planning is under-predicted to improve safety, then collision risk is reduced, but the planned path becomes too conservative and limits maneuverability
Solution Approach 1:
Real-time sensor feedback on actual steering performance allows the system to safely increase path aggressiveness when capabilities are confirmed, eliminating the need for conservative under-prediction. The system only plans paths that are verified to be within actual actuator capabilities.
Solution Approach 2:
The system performs preliminary verification of actuator capabilities through health monitoring and real-time estimation before finalizing the path plan. This preliminary assessment ensures the path is both safe and optimal, avoiding the conservatism of under-prediction while maintaining safety through pre-verification.
3Measurement precision
If real-time estimation of actuator capabilities is implemented to improve path planning accuracy, then path execution reliability is enhanced, but system complexity increases
Solution Approach 1:
The steering actuator performs multiple functions: it executes path planning commands while simultaneously providing feedback on its actual capabilities and health status. This multi-functionality eliminates the need for separate monitoring systems, reducing overall system complexity while maintaining high measurement precision.
Solution Approach 2:
The steering actuator self-monitors its own health and capability status, providing real-time feedback without requiring external diagnostic systems. This self-service capability reduces system complexity by eliminating separate monitoring components while maintaining accurate capability estimation.
4Reliability
If health monitoring of the actuator is continuously performed to improve reliability, then actuator failure is detected earlier, but computational load and processing time increase
Solution Approach 1:
The system performs partial health monitoring by focusing on the most critical actuator parameters (steering angle, velocity, acceleration) rather than comprehensively monitoring all possible failure modes. This selective monitoring maintains high reliability while minimizing computational load and processing time.
Solution Approach 2:
The system performs preliminary health assessment using pre-established models and thresholds, allowing rapid evaluation of actuator status without complex real-time analysis. This preliminary screening detects most failures quickly while reserving computational resources for detailed analysis only when needed.
Data Source
AI summary
A system comprises a processor and a memory storing instructions which when executed by the processor configure the processor to estimate a rack force for a steering system of a vehicle based on current driving and environmental conditions and estimate a health of an actuator of the steering system of the vehicle. The instructions configure the processor to estimate maximum achievable angle, velocity, and acceleration for the actuator of the steering system based on the estimated rack force and the estimated health of the actuator. The instructions configure the processor to provide to the steering system a path planned for the vehicle based on the estimated maximum achievable angle, velocity, and acceleration for the actuator of the steering system.