Yazar "Huang, Cong" seçeneğine göre listele

Listeleniyor 1 - 6 / 6
  • [ X ]
    Öğe
    A distributed state and fault estimation scheme for state-saturated systems with measurements over sensor networks
    (Elsevier, 2024) Huang, Cong; Coskun, Serdar; Karimi, Hamid Reza; Ding, Weiping
    This article explores a new framework of distributed state and fault estimation (DSFE) for the state -saturated systems over sensor networks. To this aim, the upper bound on estimation error covariance (EEC) is ensured and the explicit expression of the corresponding estimator gains is given with both quantization effects and state saturations. Further, a feasible upper bound is located on EEC and minimized by parameterizing the estimator gain. The matrix simplification technique is adopted to deal with the sensor network topology's sparseness problem. Additionally, the estimation performance is first analyzed and then ensured by conducting a sufficient condition. At last, experiments are carried out to verify the feasibility of the developed DSFE method.
  • [ X ]
    Öğe
    A multi-objective hierarchical deep reinforcement learning algorithm for connected and automated HEVs energy management
    (Pergamon-Elsevier Science Ltd, 2024) Coskun, Serdar; Yazar, Ozan; Zhang, Fengqi; Li, Lin; Huang, Cong; Karimi, Hamid Reza
    Connected and autonomous vehicles have offered unprecedented opportunities to improve fuel economy and reduce emissions of hybrid electric vehicle (HEV) in vehicular platoons. In this context, a hierarchical control strategy is put forward for connected HEVs. Firstly, we consider a deep deterministic policy gradient (DDPG) algorithm to compute the optimized vehicle speed using a trained optimal policy via vehicle-to-vehicle communication in the upper level. A multi-objective reward function is introduced, integrating vehicle fuel consumption, battery state-of-the-charge, emissions, and vehicle car-following objectives. Secondly, an adaptive equivalent consumption minimization strategy is devised to implement vehicle-level torque allocation in the platoon. Two drive cycles, HWFET and human-in-the-loop simulator driving cycles are utilized for realistic testing of the considered platoon energy management. It is shown that DDPG runs the engine more efficiently than the widely-implemented Q-learning and deep Q-network, thus showing enhanced fuel savings. Further, the contribution of this paper is to speed up the higher-level vehicular control application of deep learning algorithms in the connected and automated HEV platoon energy management applications.
  • [ X ]
    Öğe
    Actor-Critic TD3-based Deep Reinforcement Learning for Energy Management Strategy of HEV
    (Institute of Electrical and Electronics Engineers Inc., 2023) Yazar, Ozan; Coskun, Serdar; Li, Lin; Zhang, Fengqi; Huang, Cong
    In the last decade, deep reinforcement learning (DRL) algorithms have been employed in the design of energy management strategy (EMS) for hybrid electric vehicles (HEVs). Investigation of the real-time applicability of DRL algorithms as an EMS is critical in terms of training time, fuel savings, and state-of-charge (SOC) sustainability. To this end, we propose a twin delayed deep deterministic policy gradient (TD3) algorithm that is an improved version of the deep deterministic policy gradient (DDPG) algorithm for HEV fuel savings. Compared to the existing Q-learning-based reinforcement learning and the deep Q-network-based and DDPG-based deep reinforcement algorithms, the proposed TD3 provides stable training efficiency, promising fuel economy, and a lower variation range of SOC charge sustainability under various drive cycles. © 2023 IEEE.
  • [ X ]
    Öğe
    Quadratic programming-based cooperative adaptive cruise control under uncertainty via receding horizon strategy
    (Sage Publications Ltd, 2021) Coskun, Serdar; Huang, Cong; Zhang, Fengqi
    Cooperative longitudinal motion control can greatly contribute to safety, mobility, and sustainability issues in today's transportation systems. This article deals with the development of cooperative adaptive cruise control (CACC) under uncertainty using a model predictive control strategy. Specifically, uncertainties arising in the system are presented as disturbances acting in the system and measurement equations in a state-space formulation. We aim to design a predictive controller under a common goal (cooperative control) such that the equilibrium from initial condition of vehicles will remain stable under disturbances. The state estimation problem is handled by a Kalman filter and the optimal control problem is formulated by the quadratic programming method under both state and input constraints considering traffic safety, efficiency, as well as driving comfort. In the sequel, adopting the CACC system in four-vehicle platoon scenarios are tested via MATLAB/Simulink for cooperative vehicle platooning control under different disturbance realizations. Moreover, the computational effectiveness of the proposed control strategy is verified with respect to different platoon sizes for possible real-time deployment in next-generation cooperative vehicles.
  • [ X ]
    Öğe
    Robust H∞ Dynamic Output-Feedback Control for CACC With ROSSs Subject to RODAs
    (Ieee-Inst Electrical Electronics Engineers Inc, 2022) Huang, Cong; Coskun, Serdar; Wang, Jun; Mei, Peng; Shi, Quan
    This paper is concerned with the problem of dynamic output feedback control design for the cooperative adaptive cruise control (CACC) system with randomly occurring sensor saturations (ROSSs) subject to randomly occurring deception attacks (RODAs). The dynamics of the vehicle equipped with CACC system behave like a linear model in which the deviation of the position and velocity are chosen as the state variables. Sensor saturations and deception attacks are simultaneously taken into account, and two sets of Bernoulli random variables are utilized to characterize their nature of random occurrence. The main objective of this article is to develop a robust dynamic output feedback controller (RDOFC) such that, for all possible parameter uncertainties, sensor saturations as well as deception attacks, all the states can still be exponentially mean square stable and the H-infinity performance index is guaranteed. In light of the Lyapunov stability theory, a sufficient condition of the desired controller is firstly constructed, then the controller parameters are obtained relying on the solutions to a set of linear matrix inequalities (LMIs). Finally, the three-car platoon system is evaluated to validate the usefulness of the developed algorithm.
  • [ X ]
    Öğe
    State and fault estimation for nonlinear systems subject to censored measurements: A dynamic event-triggered case
    (Wiley, 2022) Huang, Cong; Coskun, Serdar; Zhang, Xin; Mei, Peng
    In this article, the state and fault estimation (SFE) issue is studied for nonlinear systems with parameter uncertainties subject to censored measurement under the dynamic event-triggering protocol. The well-known Tobit measurement model is introduced to characterize the censored measurement and the nonlinearity satisfies specific constraints. For the energy-saving purposes, a dynamic event-triggering protocol is employed to govern the measurement transmissions between the sensor and its corresponding estimator. The primary objective of the addressed problem is to design a dynamic event-triggering estimator to estimate the system state and fault signal simultaneously such that, the upper bound (UB) of the estimation error covariance is guaranteed at each iteration, then such UB is minimized by properly choosing the estimator gain. Moreover, the boundedness analysis of the designed SFE approach is conducted firstly and then a sufficient condition is presented to validate the estimation error is stable in mean-squared sense. Lastly, two experimental results are employed to validate the feasibility of the proposed estimation algorithm.