The autonomous vehicle steering system, a multi-input multi-output (MIMO) system, is challenging to design using traditional controllers due to the interaction between inputs and outputs. If PID controllers are used the control loops are executed independently of each other as there is no interaction between the loops. Designing a larger system increases the controller parameters requiring tuning. Model Predictive Control (MPC) overcomes this problem, as it is a multi-variable control method taking into account the interactions of the variables in the target system. Achieving a high safety level is also critical for autonomous vehicle systems. This can be provided by an MPC controller, which can handle constraints such as maintaining a safe distance from other cars. Wider applicability of the Model Predictive Controller calls for more efficient hardware architectures for implementation. The aim of this paper is to achieve optimal implementation of the MPC controller by increasing the computational speed in order to reduce execution time for optimization. An MPC controller is used to control the steering system of an autonomous vehicle to keep it on the desired path. A traditional MPC controller is used to control the system where the plant dynamics do not change, whereas an Adaptive MPC controller is used when the system is nonlinear or its characteristics vary with time (the longitudinal velocity changes as the vehicle moves). Results are discussed in terms of performance, resource utilization, cost, and energy-effective implementations taking into consideration a reasonable size number of constraints handled by the controller.
For the last decade, there has been great interest in studying dynamic control for unmanned aerial vehicles, but drones—although a useful technology in different areas—are prone to several issues, such as instability, the high energy consumption of batteries, and the inaccuracy of tracking targets. Different approaches have been proposed for dealing with nonlinearity issues, which represent the most important features of this system. This paper focuses on the most common control strategy, known as model predictive control (MPC), with its two branches, linear (LMPC) and nonlinear (NLMPC). The aim is to develop a model based on sensors embedded in a Tello quad-rotor used for indoor purposes. The original controller of the Tello quad-rotor is supposed to be the slave, and the designed model predictive controller was created in MATLAB. The design was imported to another embedded system, considered the master. The objective of this model is to track the reference trajectory while maintaining the stability of the system and ensuring low energy consumption. The case study in this paper compares linear and nonlinear model predictive control (MPC). The results show the efficiency of NLMPC, which provides more promising results compared to LMPC. The comparison concentrates on the energy consumption, the tracked trajectory, and the execution time. The main finding of this research is that NLMPC is a good solution to smoothly track the reference trajectory. The controller in this case processes faster, but the rotors consume more energy because of the increased values of control inputs calculated by the nonlinear controller.
Single-processor approachAs all other laws of the growth in computing, the growth of computing performance also shows a "logistic curve"-like behavior, rather than an unlimited exponential growth. The stalling of the single-processor performance experienced nearly two decades ago forced computer experts to look for alternative methods, mainly for some kind of parallelization. Solving the task needs different parallelization methods, and the wide range of those distributed systems limits the computing performance in very different ways. Some general limitations are shortly discussed, and a (by intention strongly simplified) general model of performance of parallelized systems is introduced. The model enables to highlight bottlenecks of parallelized systems of different kind and with the published performance data enables to predict performance limits of strongly parallelized systems like large scale supercomputers and neural networks. Some alternative solution possibilities of increasing computing performance are also discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.