LVD-NMPC: A Learning-based Vision Dynamics Approach to Nonlinear Model Predictive Control for Autonomous Vehicles
| Autor: | Mihai Zaha, Sorin Mihai Grigorescu, Gigel Macesanu, Bogdan Trasnea, Cosmin Ginerica |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
0209 industrial biotechnology TK7800-8360 Computer science 02 engineering and technology Systems and Control (eess.SY) Electrical Engineering and Systems Science - Systems and Control Computer Science - Robotics 020901 industrial engineering & automation Artificial Intelligence 0502 economics and business FOS: Electrical engineering electronic engineering information engineering Learning based Robot vision 050210 logistics & transportation business.industry Deep learning 05 social sciences QA75.5-76.95 Computer Science Applications Model predictive control Dynamics (music) Nonlinear model Electronic computers. Computer science Artificial intelligence Electronics business Robotics (cs.RO) Software |
| Zdroj: | International Journal of Advanced Robotic Systems, Vol 18 (2021) International Journal of Advanced Robotic Systems |
| DOI: | 10.48550/arxiv.2105.13038 |
| Popis: | In this article, we introduce a learning-based vision dynamics approach to nonlinear model predictive control (NMPC) for autonomous vehicles, coined learning-based vision dynamics (LVD) NMPC. LVD-NMPC uses an a-priori process model and a learned vision dynamics model used to calculate the dynamics of the driving scene, the controlled system’s desired state trajectory, and the weighting gains of the quadratic cost function optimized by a constrained predictive controller. The vision system is defined as a deep neural network designed to estimate the dynamics of the image scene. The input is based on historic sequences of sensory observations and vehicle states, integrated by an augmented memory component. Deep Q-learning is used to train the deep network, which once trained can also be used to calculate the desired trajectory of the vehicle. We evaluate LVD-NMPC against a baseline dynamic window approach (DWA) path planning executed using standard NMPC and against the PilotNet neural network. Performance is measured in our simulation environment GridSim, on a real-world 1:8 scaled model car as well as on a real size autonomous test vehicle and the nuScenes computer vision dataset. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|