Collision Imminent Steering Control Using Nonlinear Model Predictive Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current collision imminent steering (CIS) systems in vehicles are limited by their inability to distinguish between difficult and infeasible trajectories, often requiring conservative maneuvers that do not fully exploit the vehicle's capabilities, especially at high speeds, and struggle with aggressive lane changes due to constraints in available space.
Innovation Solution
The implementation of a nonlinear model predictive control framework that predicts and optimizes future vehicle behavior over a finite time horizon, allowing for simultaneous calculation of an optimal path and steering commands to maneuver around obstacles, using state feedback and constraints such as lane change thresholds and tire models to ensure safe and effective obstacle avoidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conservative trajectories are used to ensure the PID controller can follow the reference trajectory, then the vehicle can safely navigate obstacles, but the vehicle cannot operate at its dynamic limits and the maneuver effectiveness is reduced
Solution Approach 1:
The patent transforms the control approach by changing from a linear PID controller to a nonlinear controller that adapts parameters based on vehicle state. The controller uses feedback linearization and gains scheduling to adjust control parameters dynamically, allowing the vehicle to operate at dynamic limits while maintaining stability and reliability in trajectory following.
2Productivity
If aggressive lane change maneuvers are implemented at high speeds, then the obstacle avoidance capability is improved, but the maneuver becomes difficult to constrain and predict
Solution Approach 1:
The patent implements a dynamic control system that continuously adapts to changing vehicle states and conditions. The nonlinear controller with feedback linearization and gains scheduling adjusts control actions in real-time based on vehicle speed, steering angle, and other state variables, making aggressive maneuvers predictable and controllable while maintaining obstacle avoidance capability.
Solution Approach 2:
The patent employs extensive feedback mechanisms where the nonlinear controller continuously monitors vehicle state variables (speed, steering angle, lateral position) and adjusts control actions accordingly. This feedback loop ensures that aggressive lane change maneuvers remain predictable and controllable even at high speeds by constantly correcting deviations from the desired trajectory.
3Productivity
If the vehicle operates at dynamic limits to minimize maneuver distance, then the obstacle avoidance efficiency is improved, but the vehicle control becomes more challenging
Solution Approach 1:
The patent replaces traditional mechanical control approaches with advanced computational methods. The nonlinear controller using feedback linearization and gains scheduling substitutes simple proportional control with a sophisticated algorithm that handles vehicle nonlinearity, allowing the system to operate at dynamic limits while managing control complexity through software-based solutions.
Data Source
AI summary
A vehicle has a steering mechanism coupled to a wheel of the vehicle. The steering mechanism is adjustable to alter a vehicle trajectory. The vehicle also comprises a collision imminent steering (CIS) control system. The collision imminent steering control system includes a controller in electrical communication with the steering mechanism and is configured to adjust a steering sequence of the steering mechanism to alter the vehicle trajectory when an obstacle is detected at a distance from the vehicle less than a calculated safe braking distance. The controller simultaneously calculates a predicted optimal vehicle path around the obstacle and a steering sequence determined to follow the predicted optimal vehicle path around the obstacle using feedback received by the controller.