Browsing by Author "Chen, C. L. Philip"

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Design and implementation of deep neural network-based control for automatic parking maneuver process
    (IEEE, 2022-04-01) Chai, Runqi; Tsourdos, Antonios; Savvaris, Al; Chai, Senchun; Xia, Yuanqing; Chen, C. L. Philip
    This article focuses on the design, test, and validation of a deep neural network (DNN)-based control scheme capable of predicting optimal motion commands for autonomous ground vehicles (AGVs) during the parking maneuver process. The proposed design utilizes a multilayer structure. In the first layer, a desensitized trajectory optimization method is iteratively performed to establish a set of time-optimal parking trajectories with the consideration of noise-perturbed initial configurations. Subsequently, by using the preplanned optimal parking trajectory data set, several DNNs are trained in order to learn the functional relationship between the system state-control actions in the second layer. To obtain further improvements regarding the DNN performances, a simple yet effective data aggregation approach is designed and applied. These trained DNNs are then utilized as the motion controllers to generate feedback actions in real time. Numerical results were executed to demonstrate the effectiveness and the real-time applicability of using the proposed control scheme to plan and steer the AGV parking maneuver. Experimental results were also provided to justify the algorithm performance in real-world implementations.
  • Thumbnail Image
    ItemOpen Access
    Multiobjective overtaking maneuver planning of autonomous ground vehicles
    (IEEE, 2020-03-05) Chai, Runqi; Tsourdos, Antonios; Savvaris, Al; Chai, Senchun; Xia, Yuanqing; Chen, C. L. Philip
    This paper proposes a computational trajectory optimization framework for solving the problem of multi-objective automatic parking motion planning. Constrained automatic parking maneuver problem is usually difficult to solve because of some practical limitations and requirements. This problem becomes more challenging when multiple objectives are required to be optimized simultaneously. The designed approach employs a swarm intelligent algorithm to produce the trade-off front along the objective space. In order to enhance the local search ability of the algorithm, a gradient operation is utilized to update the solution. In addition, since the evolutionary process tends to be sensitive with respect to the flight control parameters, a novel adaptive parameter controller is designed and incorporated in the algorithm framework such that the proposed method can dynamically balance the exploitation and exploration. The performance of using the designed multi-objective strategy is validated and analyzed by performing a number of simulation and experimental studies. The results indicate that the present approach can provide reliable solutions and it can outperform other existing approaches investigated in this paper.
  • Thumbnail Image
    ItemOpen Access
    Review of advanced guidance and control algorithms for space/aerospace vehicles
    (Elsevier, 2021-03-01) Chai, Runqi; Tsourdos, Antonios; Savvaris, Al; Chai, Senchun; Xia, Yuanqing; Chen, C. L. Philip
    The design of advanced guidance and control (G&C) systems for space/aerospace vehicles has received a large amount of attention worldwide during the last few decades and will continue to be a main focus of the aerospace industry. Not surprisingly, due to the existence of various model uncertainties and environmental disturbances, robust and stochastic control-based methods have played a key role in G&C system design, and numerous effective algorithms have been successfully constructed to guide and steer the motion of space/aerospace vehicles. Apart from these stability theory-oriented techniques, in recent years, we have witnessed a growing trend of designing optimisation theory-based and artificial intelligence (AI)-based controllers for space/aerospace vehicles to meet the growing demand for better system performance. Related studies have shown that these newly developed strategies can bring many benefits from an application point of view, and they may be considered to drive the onboard decision-making system. In this paper, we provide a systematic survey of state-of-the-art algorithms that are capable of generating reliable guidance and control commands for space/aerospace vehicles. The paper first provides a brief overview of space/aerospace vehicle guidance and control problems. Following that, a broad collection of academic works concerning stability theory-based G&C methods is discussed. Some potential issues and challenges inherent in these methods are reviewed and discussed. Then, an overview is given of various recently developed optimisation theory-based methods that have the ability to produce optimal guidance and control commands, including dynamic programming-based methods, model predictive control-based methods, and other enhanced versions. The key aspects of applying these approaches, such as their main advantages and inherent challenges, are also discussed. Subsequently, a particular focus is given to recent attempts to explore the possible uses of AI techniques in connection with the optimal control of the vehicle systems. The highlights of the discussion illustrate how space/aerospace vehicle control problems may benefit from these AI models. Finally, some practical implementation considerations, together with a number of future research topics, are summarised.
  • Thumbnail Image
    ItemOpen Access
    Six-DOF spacecraft optimal trajectory planning and real-time attitude control: a deep neural network-based approach
    (IEEE, 2019-12-12) Chai, Runqi; Tsourdos, Antonios; Savvaris, Al; Chai, Senchun; Xia, Yuanqing; Chen, C. L. Philip
    This brief presents an integrated trajectory planning and attitude control framework for six-degree-of-freedom (6-DOF) hypersonic vehicle (HV) reentry flight. The proposed framework utilizes a bilevel structure incorporating desensitized trajectory optimization and deep neural network (DNN)-based control. In the upper level, a trajectory data set containing optimal system control and state trajectories is generated, while in the lower level control system, DNNs are constructed and trained using the pregenerated trajectory ensemble in order to represent the functional relationship between the optimized system states and controls. These well-trained networks are then used to produce optimal feedback actions online. A detailed simulation analysis was performed to validate the real-time applicability and the optimality of the designed bilevel framework. Moreover, a comparative analysis was also carried out between the proposed DNN-driven controller and other optimization-based techniques existing in related works. Our results verify the reliability of using the proposed bilevel design for the control of HV reentry flight in real time.