Model predictive control together with its offline computation is utilized for sparse control. Since a conventional type of sparse control is in an open-loop style, it is meaningful to realize it in a closed-loop style using the model predictive control scheme. Noting that the input of sparse control basically takes only three values, one needs to compute three sets in the state space so that membership of a state in one set implies which input value is to be given. These sets are not trivial to compute because they are neither polytopic nor convex in general. In this paper, a special coordinate transform is proposed for approximate computation of these sets with special attention on the approximation quality around the origin. The efficacy of the proposed method is illustrated by a numerical example.

Communication network formation and position control law for multiple unmanned ground vehicles (UGVs) to monitor a moving target at base station is proposed. A monitoring image of the target that acquired by nearby UGV is sent to base station through the communication network. Each UGV forms a communication network and calculates own moving amount with virtual forces using only information in local area. Since an error occurs in movement of real UGV, modified position control law is introduced. Numerical simulation and experiments show the effectiveness of this control law.

This paper proposes a feedback controller design in networked control systems, where communication channels of both sensor/controller and controller/actuator directions suffer from packet dropouts, under round-robin scheduling. We assume that the number of successive dropouts is bounded and that we are able to detect the occurrence of a packet drop without loss/delay. The proposed controller consists of an observer whose gain is switched according to the elapsed period, and feedback of estimated state as well. This enables us much less conservative design than using a single gain. The gains are designed via sufficient linear matrix inequality (LMI) conditions. The LMIs are derived by using a switched Lyapunov function method. The effectiveness of the proposed method is illustrated by numerical simulations.

In this paper, we propose a new update scheme for the controller parameters of the two-degree-of-freedom control system using Virtual Internal Model Tuning. Here, both of the feedback controller and the feed-forward one are simultaneously updated to improve the closed loop property and the tracking property, respectively. The key of the proposed technique is to use the virtual model, which is derived from the ideal controller for a given specification. Finally, to show the usefulness of the proposed method, verification is performed using numerical simulation.

This paper proposes a novel cyber-physical traffic signal control system that interconnects physical dynamics and an optimization algorithm in order to reduce the street congestion. We first present a macroscopic traffic flow model that describes time evolution of the vehicle numbers on each road. Using this model, we then formulate an optimization problem to optimize the steady states/inputs in the sense of minimizing the congestion. We then present a (distributed) solution to the problem based on so-called partial primal-dual gradient algorithm. The traffic flow model is then shown to be successfully embedded in the algorithm as a sub-process for optimization, and accordingly a novel cyber-physical system is obtained. We then rigorously prove asymptotic optimality for constant disturbances and input-output stability for time-varying ones. Finally, we demonstrate the present algorithm through simulation on a microscopic traffic simulator, UC-win/Road, capable of the real-time feedback. It is then confirmed that the value of objective function for the optimization problem is reduced by 50% as compared to a fixed-time signal control scheme.

This paper describes the application of Monte Carlo model predictive control (MCMPC) to systems with collision phenomena. In general, a physical quantity related to velocity discontinuously changes at the moment of a collision. Gradient-based model predictive control that requires gradient information of the cost function can not handle the collision phenomena directly. On the other hand, MCMPC can calculate local optimal solutions and stationary solutions needed each time as far as the forward simulation is feasible. Therefore, we examined the problem of swing up and stabilizing the cart type inverted pendulum with the collision of the cart. If the controller can properly handle the discontinuity, faster swing-up may be realized by exploiting impact force caused by the collision of a cart within the evaluation function. In this paper, we show the result of simulations and real experiments, and report that we can realize speeding up of swing-up using collision.

With improving navigation systems, commercial aircraft can take more flexible routes. However, routes of the aircraft still have several operational restrictions, such as sector boundaries. This paper shows the trajectory optimization method for arrival aircraft considering the constrains due to the sector boundaries. The obstacle layout method is employed to set the constraints. To express the sector boundaries, the obstacles are set to the trajectory optimization by referring to radar data. Numerical simulations are conducted to show that the trajectory optimization method with the obstacle layout method can calculate the optimal arrival trajectory considering the constraints due to the sector boundaries.

In a modern aging society, electric wheelchairs are an essential means of transportation for the elderly. However, human operation errors of the electric wheelchair cause some accidents. For a control system operated by a human, human assist control to avoid accidents attracts much attention in recent years. For this problem, Nakamura et al. have proposed a relaxed control barrier function and a human assist control law. However, considering the operability and safety, it was difficult to apply the proposed human assist control to an electric wheelchair. In this paper, we propose a novel human assist control law that can consider an assist profile function based on a relaxed CBF. We aim to design human assist control law that can be applied to electric wheelchairs. The effectiveness of the proposed control method is confirmed by simulations and actual experiments using an electric wheelchair.

Feedback error learning (FEL) control attains accurate response to a target signal by tuning parameters in the feedforward (FF) controller, provided that feedback (FB) control stabilizes closed-loop in two-degree-of-freedom structure. The authors proposed an FEL tuning law by which the output error converges to zero for any target signals, under a strictly positive real (SPR) condition of the closed-loop. However, this condition is not always satisfied even if the plant is biproper or has a relative degree one. In this paper, we propose to satisfy the SPR condition by designing a filter in the FF controller by solving i) a linear matrix inequality (LMI) for a nominal plant; and ii) a finite set of LMIs for the uncertain plant in a polytope representation. We verify the effectiveness of these methods via a numerical example.

In the construction industry, skilled operators of hydraulic excavators have been decreasing, and decline in productivity at the site is concerned. Especially, operation skills of hydraulic excavators are important to obtain high productivity. In the case of skilled operator, smooth and superior motions are achieved by appropriate operations. However, it is difficult to achieve these motions for non-experts. In this paper, a control system which achieves smooth and highly workable operation even for a non-expert is newly proposed. The Center of Mass (CoM) velocity of attachment during digging operation is controlled by the proposed controller whose parameters are tuned by the Fictitious Reference Iterative Tuning (FRIT) method.

In motion planning of robots such as quadrotors, potential field methods are useful so that robots avoid obstacles. The artificial potential field method, which is one of the potential field ones, enables us to plan actions. However, the quadrotors sometimes fail to avoid the obstacles because the artificial potential field method does not take into consideration the inertia effect arising from the velocity of the quadrotors. To overcome the inertia effect, we give an idea of applying deep reinforcement learning to the artificial potential field method to determine an additional reference signal to the quadrotor. Thanks to this reference signal, the quadrotor improves the performance in trial and error to avoid the obstacles. Then the robot achieves an optimal action from the velocity of the robot and the position of the obstacles.

This paper describes a tuning method of a rehabilitation system for tongue exercise such that participants control a cursor which moves to the target on the display by the movement of the tongue. As a rehabilitative training, each participant has different movement capability, therefore operational feelings and training loads are to be different. Therefore, we propose a method for adjusting the trajectories among participants. The time base generator (TBG), which is a bio-mimetic trajectory, is employed as the desired response of the cursor trajectories and a fictitious reference iterative tuning (FRIT) approach is used for tuning the system. Finally, simulations show the effectiveness of the proposed method.

In this paper, a model-based control system for a diesel engine was evaluated under random accelerator operation. The control system consists of two-degree-of—freedom controllers of air path system and combustion system. For the air path system, a physical model was developed and a feedforward (FF) controller was developed by an inverse model and virtual feedback (FB) loop. An FB controller was designed based on H_∞ control theory. For the combustion system, a physical model was developed and its inverse model was used as an FF controller. An FB controller was designed based on an adaptive output feedback theory. The overall control system was implemented in a rapid prototyping system. For the evaluation of controller performance, vehicle dynamics simulation and a real engine were used and accelerator position was changed randomly. The model-based control system showed good performance.