Combination Of Model Predictive Control Research Articles

Fault-tolerant control for satellite autonomous rendezvous with quality characteristics and actuator uncertainties

An autonomous fault detection and fault-tolerant control solution combining Model Predictive Control (MPC) and Label Distributed Learning (LDL) is developed for the position tracking and attitude synchronization problems in autonomous satellite rendezvous and docking. The proposed algorithm can deal with center of mass (CM) variations and time-varying mass characteristics due to satellite fuel consumption, actuator saturation and fault, and external disturbances. Firstly, a six-degree-of-freedom (6-DOF) satellite translation-rotation coupling model was established to express the relative motion between the satellite and the target, taking the above factors into account. Subsequently, an on-board autonomous closed-loop fault-tolerant control algorithm using MPC is proposed, in which an LDL-based fault detection mechanism is embedded. The scheme specifies that propulsion faults are estimated by calculating the difference between the MPC's prediction of the satellite's future state and the measured change in the actual state. And the mapping matrix between the commanded force/torque and the actual force of the propulsion system is updated in real time to achieve autonomous fault detection and compensation on board the satellite. Among them, the LDL-based fault detection scheme is trained offline. It requires only simple matrix operations for on-orbit use, which is low-cost and suitable for resource-limited space environments. The additional creation of a time-varying disturbance observer allows the estimation and compensation of unmodelled errors and external disturbances. The stability of the proposed fault-tolerant control algorithm is rigorously demonstrated using Lyapunov analysis. Finally, numerical simulations show that the proposed method can successfully achieve autonomous satellite docking under different combinations of thruster faults.

High‐level decision‐making for autonomous overtaking: An MPC‐based switching control approach

AbstractThe key motivation of this paper lies in the development of a high‐level decision‐making framework for autonomous overtaking maneuvers on two‐lane country roads with dynamic oncoming traffic. To generate an optimal and safe decision sequence for such scenario, an innovative high‐level decision‐making framework that combines model predictive control (MPC) and switching control methodologies is introduced. Specifically, the autonomous vehicle is abstracted and modelled as a switched system. This abstraction allows vehicle to operate in different modes corresponding to different high‐level decisions. It establishes a crucial connection between high‐level decision‐making and low‐level behaviour of the autonomous vehicle. Furthermore, barrier functions and predictive models that account for the relationship between the autonomous vehicle and oncoming traffic are incorporated. This technique enables us to guarantee the satisfaction of constraints, while also assessing performance within a prediction horizon. By repeatedly solving the online constrained optimization problems, we not only generate an optimal decision sequence for overtaking safely and efficiently but also enhance the adaptability and robustness. This adaptability allows the system to respond effectively to potential changes and unexpected events. Finally, the performance of the proposed MPC framework is demonstrated via simulations of four driving scenarios, which shows that it can handle multiple behaviours.

Optimization Method of Multi-Mode Model Predictive Control for Wind Farm Reactive Power

This paper presents a novel approach for optimizing wind farm control through the utilization of a combined model predictive control method. In contrast to conventional methods of controlling active and reactive power in wind farms, the suggested approach integrates a wind power prediction model driven by a neural network and a state-space model for wind turbines. This combination facilitates a more precise forecast of active power, thereby enabling the dynamic prediction of the range of reactive power output from the wind turbines. When combined with the equation of state in wind farm space, it is possible to accurately optimize the reactive power of a wind farm. Furthermore, the impact of active power on voltage fluctuations in the wind farm collector system was examined. The utilization of model predictive control enhances voltage regulation, optimizes system redundancy, and increases the reactive capacity. Sensitivity coefficients were calculated using analytical methods to enhance computational efficiency and to resolve issues related to convergence. In order to validate the proposed methodology and control scheme, a wind farm simulation model comprising 20 turbines was developed to assess the feasibility of the scheme.

Longitudinal and lateral stability control for autonomous vehicles in curved road scenarios with road undulation

PurposeThis research aims to present a novel cooperative control architecture designed specifically for roads with variations in height and curvature. The primary objective is to enhance the longitudinal and lateral tracking accuracy of the vehicle.Design/methodology/approachIn addressing the challenges posed by time-varying road information and vehicle dynamics parameters, a combination of model predictive control (MPC) and active disturbance rejection control (ADRC) is employed in this study. A coupled controller based on the authors’ model was developed by utilizing the capabilities of MPC and ADRC. Emphasis is placed on the ramifications of road undulations and changes in curvature concerning control effectiveness. Recognizing these factors as disturbances, measures are taken to offset their influences within the system. Load transfer due to variations in road parameters has been considered and integrated into the design of the authors’ synergistic architecture.FindingsThe framework's efficacy is validated through hardware-in-the-loop simulation. Experimental results show that the integrated controller is more robust than conventional MPC and PID controllers. Consequently, the integrated controller improves the vehicle's driving stability and safety.Originality/valueThe proposed coupled control strategy notably enhances vehicle stability and reduces slip concerns. A tailored model is introduced integrating a control strategy based on MPC and ADRC which takes into account vertical and longitudinal force variations and allowing it to effectively cope with complex scenarios and multifaceted constraints problems.

Cautious Bayesian MPC: Regret Analysis and Bounds on the Number of Unsafe Learning Episodes

This paper investigates the combination of model predictive control (MPC) concepts and posterior sampling techniques and proposes a simple constraint tightening technique to introduce cautiousness during explorative learning episodes. The provided theoretical analysis in terms of cumulative regret focuses on previously stated sufficient conditions of the resulting ‘Cautious Bayesian MPC’ algorithm and shows Lipschitz continuity of the future reward function in the case of linear MPC problems. In the case of nonlinear MPC problems, it is shown that commonly required assumptions for nonlinear MPC optimization techniques provide sufficient criteria for model-based RL using posterior sampling. Furthermore, it is shown that the proposed constraint tightening implies a bound on the expected number of unsafe learning episodes in the linear and nonlinear case using a soft-constrained MPC formulation. The efficiency of the method is illustrated using numerical examples.

Path planning algorithm in complex environment based on DDPG and MPC

The problem of mixed static and dynamic obstacle avoidance is essential for path planning in highly dynamic environments. However, the paths formed by grid edges can be longer than the actual shortest paths in the terrain since their headings are artificially constrained. Existing methods can hardly deal with dynamic obstacles. To address this problem, we propose a new algorithm combining Model Predictive Control (MPC) with Deep Deterministic Policy Gradient (DDPG). Firstly, we apply the MPC algorithm to predict the trajectory of dynamic obstacles. Secondly, the DDPG with continuous action space is designed to provide learning and autonomous decision-making capability for robots. Finally, we introduce the idea of the Artificial Potential Field to set the reward function to improve convergence speed and accuracy. In this paper, Matplotlib is used for simulation experiments. The results show that our method has improved considerably in accuracy by 8.11% -43.20% compared with theother methods, and on the length of the path and turning angle by reducing about 50 units and 340 degrees compared with DeepQ Network (DQN), respectively. We also employ Unity 3D to perform simulation experiments in highly uncertain environments, such as squares.

Research on Truck Lane Management Strategies for Platooning Speed Optimization and Control on Multi-Lane Highways

Automated truck platooning has become an increasingly popular research subject, and its applicability to highways is considered one of the earliest possible landing scenarios for automated driving. However, there is a lack of research regarding the combination of truck platooning technology and truck lane management strategy on multilane highways in the environment of a cooperative vehicle–infrastructure system (CVIS). For highway weaving sections under the CVIS environment, this paper proposes a truck platooning optimal speed control model based on multi-objective optimization. Through a combination of model predictive control and the cell transmission model, this approach considers the bottleneck cell traffic flow, overall vehicle travel time, and truck platooning fuel consumption as objectives. By conducting experiments on a mixed traffic flow simulation platform, the multi-lane management strategies and optimal speed control effect were evaluated through different scenarios. This study also determined the appropriate proportion of truck platooning for an exclusive lane and to increase truck lanes, thus providing effective lane management decision support for highway managers.

Design and assessment of a core-power controller for lithium-cooled space nuclear reactor based on the concept of fuzzy model predictive control

Thanks to its unique characteristics of high power-to-mass ratio, shallow reactivity poisoning, and quick response to reactivity control, power supply system based on lithium-cooled space nuclear reactor is preferred for various exploration missions into outer and deep space. However, due to its nature of few-people or even unmanned on-duty, an intelligent autonomous control of the reactor system, especially an accurate control of the reactor core power following the demanding power output, is of vital importance for such a space nuclear reactor. In this study, a core-power controller for a megawatt ultra-small lithium-cooled space nuclear reactor was designed based on the concept of fuzzy model predictive control (FMPC) combining model predictive control and T-S fuzzy theory. Performance of the FMPC controller was simulated and assessed with the Simulink platform for five typical operation transients including ramp, step and disturbance transient. The results show that the intelligent FMPC controller possesses an excellent load-following ability and anti-interference ability, both of which are of vital importance for space exploration missions. When compared with the classical PID controller, the FMPC controller designed in this study shows also a much better performance with smaller overshoot, lesser adjusting time and lower integral time-squared error.

Hierarchical Vessel Autonomous Operation in a Port with Safety Certificates: Combined MPC and CBF Approach

In this paper, we investigate autonomous vessel operation in/near a port and present a novel hierarchical control architecture that combines model predictive control (MPC) and control barrier function (CBF). Our architecture is composed of two layers, navigation and control, and switches control schemes as the ideal vessel operations vary depending on the vessel locations. Specifically, we divide the operation into three phases, namely approaching phase, breakwater passing phase, and docking phase. In all phases, we employ the CBF-based online optimization with a vessel kinematic model at a lower layer to certify safety even in the presence of the prediction error in the navigation layer. The navigation layer is designed based on MPC so as to smoothen the collision avoidance behavior and to reflect various specifications stemming from laws. The present control architecture is demonstrated through various simulations including the one with real data of vessels in Tokyo Bay.

Active Fault-Tolerant Control Based on MPC and Reinforcement Learning for Quadcopter with Actuator Faults

In this paper, we propose an active fault-tolerant control (AFTC) method combining model predictive control (MPC) and reinforcement learning (RL) for the quadcopter with actuator faults. We take a data-based discriminant model as the fault detection and diagnosis (FDD) module indicating a system fault mode based on the state error. With the information of the fault mode and the state error, the RL controller generates auxiliary control signals to correct the system. To configure the MPC controller quickly, we propose an auxiliary signal-based method for estimation of fault parameters and prove its convergence. The AFTC framework reduces requirements for accurate modeling, and avoids the instability of the RL controller under a continuous operation. To validate the effectiveness of the proposed framework, two trajectory tracking simulations with single and multiple faults are carried out. The simulation results show satisfactory performance and verify that the proposed framework is real-time applicable.

Stability Study of Time Lag Disturbance in an Automatic Tractor Steering System Based on Sliding Mode Predictive Control

To improve the working accuracy and anti-interference capability of the steering operation of an automatic tractor, this paper investigates the tractor steering system. In response to the current problems of high steering resistance during tractor field operations and the low service life of the drive shaft of conventional electric steering wheel solutions, an electro-hydraulic coupled power-assisted solution combining EPS and HPS is proposed. The combination of the EPS system’s high control accuracy and sensitive steering operation and the hydraulic power system’s large steering torque greatly reduces the power of the power motor and battery performance requirements, optimizing the power transmission scheme to achieve green and energy-saving purposes. Secondly, the research is focused on the influence of external disturbances on the stability of the steering system during tractor operation, and a combination of model predictive control and sliding mode control is used to study the steering system control strategy. It is finally demonstrated through simulations and experiments that it can compensate for the disturbance of the system control parameters by external disturbances, has the ability of MPC to handle input constraints and maintains the advantages of SMC robustness.

Squat motion of a bipedal robot using real‐time kinematic prediction and whole‐body control

AbstractSquatting is a basic movement of bipedal robots, which is essential in robotic actions like jumping or picking up objects. Due to the intrinsic complex dynamics of bipedal robots, perfect squatting motion requires high‐performance motion planning and control algorithms. The standard academic solution combines model predictive control (MPC) with whole‐body control (WBC), which is usually computationally expensive and difficult to implement on practical robots with limited computing resources. The real‐time kinematic prediction (RKP) method is proposed, which considers upcoming reference motion trajectories and combines it with quadratic programming (QP)‐based WBC. Since the WBC handles the full robot dynamics and various constraints, the RKP only needs to adopt the linear kinematics in the robot's task space and to softly constrain the desired accelerations. Then, the computational cost of derived closed‐form RKP is greatly reduced. The RKP method is verified in simulation on a heavy‐loaded bipedal robot. The robot makes rapid and large‐amplitude squatting motions, which require close‐to‐limit torque outputs. Compared with the conventional QP‐based WBC method, the proposed method exhibits high adaptability to rough planning, which implies much less user interference in the robot's motion planning. Furthermore, like the MPC, the proposed method can prepare for upcoming motions in advance but requires much less computation time.

Intelligent air defense task assignment based on hierarchical reinforcement learning.

Modern air defense battlefield situations are complex and varied, requiring high-speed computing capabilities and real-time situational processing for task assignment. Current methods struggle to balance the quality and speed of assignment strategies. This paper proposes a hierarchical reinforcement learning architecture for ground-to-air confrontation (HRL-GC) and an algorithm combining model predictive control with proximal policy optimization (MPC-PPO), which effectively combines the advantages of centralized and distributed approaches. To improve training efficiency while ensuring the quality of the final decision. In a large-scale area air defense scenario, this paper validates the effectiveness and superiority of the HRL-GC architecture and MPC-PPO algorithm, proving that the method can meet the needs of large-scale air defense task assignment in terms of quality and speed.

Predictive energy management with engine switching control for hybrid electric vehicle via ADMM

This paper studies energy management (EM) of a power-split hybrid electric vehicle (HEV) equipped with planetary gear sets. We first formulate a mixed-integer global optimal control problem that includes a binary switching variable. Convex modeling, including the fuel model for a compound energy conversion unit, is then presented to reformulate the mixed-integer EM as a two-step program. For optimizing the engine switching and battery power decisions in the first step, we employ the alternating direction method of multipliers (ADMM) algorithm where the solution of the convex relaxation is used to initialize the non-convex problem. On the standard driving cycle, simulation results indicate that the ADMM based EM method saves 7.63% fuel compared to a heuristic method, and shows 99% optimality compared to a dynamic programming method, while saving three orders of magnitude in computing time. An ADMM combined model predictive control (ADMM-MPC) method is also developed that is suitable for receding horizon control scenarios. The ADMM-MPC method shows 5.28% fuel saving when implemented using a prediction horizon of 15 samples. Meanwhile, the mean computing time for MPC updates is 3.53ms. Our results demonstrate that the proposed ADMM is capable of real-time control.

Pruning deep neural networks generates a sparse, bio-inspired nonlinear controller for insect flight.

Insect flight is a strongly nonlinear and actuated dynamical system. As such, strategies for understanding its control have typically relied on either model-based methods or linearizations thereof. Here we develop a framework that combines model predictive control on an established flight dynamics model and deep neural networks (DNN) to create an efficient method for solving the inverse problem of flight control. We turn to natural systems for inspiration since they inherently demonstrate network pruning with the consequence of yielding more efficient networks for a specific set of tasks. This bio-inspired approach allows us to leverage network pruning to optimally sparsify a DNN architecture in order to perform flight tasks with as few neural connections as possible, however, there are limits to sparsification. Specifically, as the number of connections falls below a critical threshold, flight performance drops considerably. We develop sparsification paradigms and explore their limits for control tasks. Monte Carlo simulations also quantify the statistical distribution of network weights during pruning given initial random weights of the DNNs. We demonstrate that on average, the network can be pruned to retain a small amount of original network weights and still perform comparably to its fully-connected counterpart. The relative number of remaining weights, however, is highly dependent on the initial architecture and size of the network. Overall, this work shows that sparsely connected DNNs are capable of predicting the forces required to follow flight trajectories. Additionally, sparsification has sharp performance limits.

Path tracking control strategy for off-road 4WS4WD vehicle based on robust model predictive control

With the increasing requirements for vehicle adaptability and maneuverability in various road environments, four-wheel-steering four-wheel-drive (4WS4WD) vehicles have attracted wider attention. This paper presents a robust model predictive control-based strategy for the path tracking of 4WS4WD vehicles considering external disturbances. The strategy combines model predictive control (MPC) and control allocation under an upper–lower structure. The main objective of the present work is to improve the robustness and stability of path tracking by developing an MPC algorithm in the upper layer. The controller design considers general disturbances caused by allocation errors and sudden disturbances caused by an outer force in the offset model. Based on the offset model, a robust MPC control law is obtained by converting the robustness constraints into a linear matrix inequality. The control law is mathematically demonstrated to be stable in multidisturbed conditions via the Lyapunov stability theorem. Through comparison with a similar control algorithm of path tracking and applying it on different uneven ground conditions, the proposed robust algorithm is found to effectively overcome disturbances on the system.

Combined model predictive control and ANN-based forecasters for jointly acting renewable self-consumers: An environmental and economical evaluation

Recent European Community directives introduce Renewable Energy Communities (REC) and Jointly Acting Renewable Self-Consumers (JARSC). Both entities are constituted by communities of residential and/or non-residential prosumers, located in proximity of renewable generators and Electrical Storage Systems (ESS) owned and managed by the REC/JARSCs. These aggregations of prosumers are aimed at providing environmental and economic benefits by maximizing their global self-consumption. In this frame, it is relevant to introduce a control strategy which considers the whole system represented by the REC/JARSCs and performs optimal management of energy production, storage and consumption. The present paper proposes a Model Predictive Control (MPC) based control design, targeted at the minimization of electricity cost and equivalent CO2 emissions, considering the whole ensemble of loads included in the REC/JARSCs over a 24-h prediction horizon. To exploit the MPC ability of including forecasts in the optimization problem, predictors including Artificial Neural Networks (ANN) are developed for solar irradiance, air temperature, electricity price and carbon intensity. The proposed control performance is evaluated considering a case study located in Milan, Italy, and its advantages with respect to traditional control algorithms are highlighted by comprehensive numerical simulations. Lastly, an economic evaluation of the considered system is presented.

Semi-active switching vibration control with tree-based prediction and optimization strategy

A novel control strategy that combines model predictive control (MPC) with semi-active vibration control that uses a highly effective, energy-efficient, and stable piezoelectric transducer is proposed in this paper. Incorporating MPC into semi-active vibration control enables significant improvements in control performance and robustness. However, it is challenging to directly predict and optimize the input trajectory because the semi-active input has a state-dependent discontinuous nature. To realize effective optimal control, we need a strategy that can predict the discontinuous semi-active input trajectory in a reasonable manner and is computationally cost-efficient. The proposed method employs a prediction algorithm based on a tree data structure. The proposed algorithm achieves flexible prediction and optimization of a semi-active input trajectory with a simple tree traversal. In addition, the proposed method employs a switching criterion to minimize the computational cost and implement fast prediction and optimization. The proposed method is called predictive switching vibration control with tree-based formulation and optimization, or the PSTFO method. The simulation proved that the proposed PSTFO method can predict discontinuous semi-active input and realizes optimal vibration control performance and high robustness. In addition, the high control performance and robustness of the proposed method were experimentally validated.

Robust controller for grid‐tied PV inverters based on continuous non‐linear predictive and integral sliding mode control

This paper proposes a robust continuous nonlinear control method for grid-tied photovoltaic (PV) inverters by combining model predictive control and integral sliding mode control (ISMC). The robustness of the proposed augmented controller is demonstrated mathematically by showing that the equivalent control signal of the ISMC system cancels disturbances to the system. Further, the proposed controller can compensate for system perturbations without knowing their characteristics, thereby simplifying controller design for the PV systems. The robustness of the proposed controller against parameter uncertainties and external disturbances was demonstrated via experimental and simulation verification.

Online model-based reinforcement learning for decision-making in long distance routes

In road transportation, long-distance routes require scheduled driving times, breaks, and rest periods, in compliance with the regulations on working conditions for truck drivers, while ensuring goods are delivered within the time windows of each customer. However, routes are subject to uncertain travel and service times, and incidents may cause additional delays, making predefined schedules ineffective in many real-life situations. This paper presents a reinforcement learning (RL) algorithm capable of making en-route decisions regarding driving times, breaks, and rest periods, under uncertain conditions. Our proposal aims at maximizing the likelihood of on-time delivery while complying with drivers’ work regulations. We use an online model-based RL strategy that needs no prior training and is more flexible than model-free RL approaches, where the agent must be trained offline before making online decisions. Our proposal combines model predictive control with a rollout strategy and Monte Carlo tree search. At each decision stage, our algorithm anticipates the consequences of all the possible decisions in a number of future stages (the lookahead horizon), and then uses a base policy to generate a sequence of decisions beyond the lookahead horizon. This base policy could be, for example, a set of decision rules based on the experience and expertise of the transportation company covering the routes. Our numerical results show that the policy obtained using our algorithm outperforms not only the base policy (up to 83%), but also a policy obtained offline using deep Q networks (DQN), a state-of-the-art, model-free RL algorithm.