A Physics-Driven Artificial Agent for Online Time-Optimal Vehicle Motion Planning and Control

Abstract

This paper presents a hierarchical framework with novel analytical and neural physics-driven models, to enable the online planning and tracking of minimum-time maneuvers, for a vehicle with partiallyunknown parameters. We introduce a lateral speed prediction model for high-level motion planning with economic nonlinear model predictive control (E-NMPC). A low-level steering controller is developed with a novel feedforward-feedback physics-driven artificial neural network (NN). A longitudinal dynamic model is identified to tune a low-level speed-tracking controller. The high-and low-level control models are identified with an automatic three-step scheme, combining open-loop and closed-loop maneuvers to model the maximum acceleration G-G-v performance constraint for E-NMPC, and to capture the effect of the longitudinal acceleration on the lateral dynamics. The proposed framework is used in a simulation environment, for the online closed-loop control of a highly detailed sedan vehicle simulator, whose parameters are partiallyunknown. Two different circuits are adopted to validate the approach, and a robustness analysis is performed by varying the vehicle mass and the load distribution. A minimum-time optimal control problem is solved offline and used for a comparison with the closed-loop results. A video demonstrating both the automatic three-step identification scheme and the motion planning and control results is available at the following link: https://www.youtube.com/watch?v=xQ_T96IjGP8. INDEX TERMSAutonomous racing, model learning, model predictive control (MPC), motion planning, neural networks, trajectory optimization. Open Access funding provided by 'Università degli Studi di Trento' within the CRUI CARE Agreement

Multi-Layered Local Dynamic Map for a Connected and Automated In-Vehicle System

“…In fact, thanks to entity positioning, match detection, data persistence, and action prediction algorithms, we could easily develop the collision avoidance algorithm with just some additional logic and a few lines of code. That is why we are planning to embed our LDM in our motion planning framework for at-limit handling of racing vehicles [41], to have a separate application that can take the burden of analysing the environment of interest and serve vital information quickly, reliably, and only when needed (e.g., an incoming collision or the next action of an opponent on the racetrack).…”

“…• Enhancing the matching algorithm by varying parameters based on the type of entity detected; • Converting the LDM to C++ and using the igraph package [40] to manage the graph database directly in the API; • Using the LDM to augment the capabilities of our motion planning framework for at-limit handling of racing vehicles [41]; • Using additional advanced estimation techniques [39] to increase the accuracy of the GNSS system; • Deploying the developed solution on the prototype Automated Vehicle from CRF Trento Branch and testing it online.…”

Planning Speed Mode of All-Wheel Drive Autonomous Vehicles Considering Complete Constraint Set

“…Dynamic models can be both complex and simple to control a single aspect. For example, in [6] a longitudinal dynamic model is identified to tune a low-level speedtracking controller. In [7] a simplified constrained model uses a second-order integrator to match the feasible AV dynamic behavior.…”

“…Another popular class is the LQR-type controllers. For increasing performance and overcoming physical limitations, a technique can be represented by the augmented Lagrangian framework [6], which refers to iterative LQR (ILQR) and Constrained Iterative LQR (CILQR), respectively. In [12], the sliding mode control (SMC) calculates the total driving force for longitudinal control.…”

“…In [9], the Nonlinear Model Predictive Control (NMPC) strategy was aimed at controlling a small-scale car model for autonomous racing competitions. Study [6] presented a hierarchical framework with neural physics-driven models to enable the online planning and tracking of minimum-time maneuvers. A lateral speed prediction model for high-level motion planning was considered with economic nonlinear model predictive control (E-NMPC).…”