Collision Imminent Steering Using Nonlinear Model Predictive Control
| Autor: | Tulga Ersal, Jeffrey L. Stein, John Wurts |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0209 industrial biotechnology
Computer science Active safety 020302 automobile design & engineering 02 engineering and technology Slip (materials science) Optimal control Collision Vehicle dynamics Model predictive control 020901 industrial engineering & automation 0203 mechanical engineering Control theory Obstacle Steering system Slip angle Slip (vehicle dynamics) |
| Zdroj: | ACC |
| Popis: | Collision imminent steering is an automotive active safety feature designed to swerve and avoid an obstacle on the road if the vehicle detects a forward collision cannot be avoided by braking alone. A collision imminent steering system is formulated in this paper to generate an optimal control sequence that performs an aggressive lane change maneuver in the shortest distance possible in a highway driving environment. The algorithm uses nonlinear model predictive control that pushes the vehicle to its dynamic handling limits, which are established through limits on the tire slip angle. Additionally, the algorithm is developed to leverage active four wheel steering to improve the performance over traditional front only steering. Numerical results show a window exists for which the vehicle can change lanes safely, even at highway speed. A Pareto front is generated to quantify the trade-off between maximum allowable tire slip and safe steering distance. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|