The lateral dynamics of the vehicle is strongly impacted by the longitudinal speed bringing problems to trajectory tracking. In this paper, several control structures based on PID are compared for the lateral guidance of autonomous vehicle. It includes a classic PID, a nested loop and a multi-PID controller. A comparison is made through an overtaking simulation for different speeds. The paper highlights that a PID alone cannot guarantee the stability of the closed loop in every situation and that among the three structures, the multi-PID controller allows the best tracking.
An elegant way to exploit control redundancies available in multilevel converters is through the formulation of a constrained optimization problem. An interesting connection can be made with the so-called control allocation problems defined in over-actuated constrained systems. Redundancies and constraints are taken into account to achieve the best performance. In this paper, we introduce the first result of an investigation of control allocation methods for multilevel conversion. The method is dedicated to the flying-capacitor inverter with focus on the active balancing of capacitor voltages to ensure admissible blocking voltages for the switches. A linear program is formulated and solved by using the well-established simplex algorithm. Fast variations of the DC-bus voltage were applied in simulations. Disturbances are well rejected thanks to a highly reactive balancing, and a safe switching operation is ensured.
In cloud computing management, the dynamic adaptation of computing resource allocations under timevarying workload is an active domain of investigation. Several control strategies were already proposed. Here the modelfree control setting and the corresponding “intelligent” controllers, which are most successful in many concrete engineering situations, are employed for the “horizontal elasticity.” When compared to the commercial “Auto-Scaling” algorithms, our easily implementable approach, behaves better even with sharp workload fluctuations. This is confirmed by experiments
on Amazon Web Services (AWS).
In a context of Connected Autonomous Vehicle (CAV), this paper proposes a generic hierarchical architecture for Global Chassis Control (GCC). The four levels of this architecture, namely : the Supervisor, the Global Control, the Allocation and the Local Control, are detailed. In a second part, an example whose objective is to illustrate the design process of such a hierarchical architecture is presented. For didactic reasons, and without that it doesn’t harm the general approach, the field of
study of this example is summed up to a braking situation in a straight line on dry, smooth and horizontal road, the objective being to hold the chassis under driving sollicitations in the operating field associated with comfort. A Robust Command Frequency Synthesis (RCFS) based on the CRONE control is applied for global and local control. The comparison of the simulated time
performances using a model with 14 degrees of freedom of the device in both active and in degraded mode clearly highlights the interest of the proposed approach.
Path planning is an essential stage for mobile robot control. It is more newsworthy than ever in the automotive context and especially for autonomous vehicle. Also, path planning methods need to be adaptive regarding life situations, traffic and obstacle crossing. In this paper, potential field methods are proposed to cope with these constraints and autonomous vehicles are considered equipped with all necessary sensors for obstacle detection. In this way, Ge&Cui’s attractive potential field and fractional attractive potential field have been adapted to the context of autonomous vehicles.
In this way, this latter method ensures better stability degree robustness with controlled vehicle acceleration.
In the last few years much effort has been made towards more autonomous vehicles and fuel consumption reduction. This article deals with the issue trajectory optimization of unmanned terrestrial vehicles so as to reduce consumption, travel time or to improve comfort. Main focuses are set on testing different criteria and the possibility of using a genetic algorithm to improve the potential field methods. The main idea of this article is that potential field methods could be improved by adding a dynamic target in it. Two improvements are brought to the potential field method : the generation of an optimal path in the environment, and the joint generation an optimal motion.
In today’s literature Model-Free Control, or MFC, and Active Disturbance Rejection Control, or ADRC, are
the most prominent approaches in order to keep the benefits of PID controllers, that are so popular in the industrial world,
and in the same time for attenuating their severe shortcomings. After a brief review of MFC and ADRC, several examples
show the superiority of MFC, which permits to tackle most easily a much wider class of systems.