Senchun Chai

Improved Gradient-Based Algorithm for Solving Aeroassisted Vehicle Trajectory Optimization Problems

The Space Maneuver Vehicles (SMV) [1, 2] will play an increasingly important role in the future exploration of space, since their on-orbit maneuverability can greatly increase the operational flexibility and are more difficult as a target to be tracked and intercepted. Therefore, a well-designed trajectory, particularly in skip entry phase, is a key for stable flight and for improved guidance control of the vehicle [3, 4]. Trajectory design for space vehicles can be treated as an optimal control problem. Due to the high nonlinear characteristics and strict path constraints of the problem, direct methods are usually applied to calculate the optimal trajectories, such as direct multiple shooting method [5], direct collocation method [5, 6], or hp-adaptive pseudospectral method [7, 8]. Nevertheless, all the direct methods aim to transcribe the continuous-time optimal control problems to a Nonlinear Programming Problem (NLP). The resulting NLP can be solved numerically by well-developed algorithms such as Sequential Quadratic Programming (SQP) and Interior Point method (IP) [9, 10]. SQP methods are used successfully for the solution of large scale NLPs. Each Newton iteration of the SQP requires the solution of a quadratic programming subproblem

Stability of a Class of Multiagent Tracking Systems With Unstable Subsystems

In this paper, we predeploy a large number of smart agents to monitor an area of interest. This area could be divided into many Voronoi cells by using the knowledge of Voronoi diagram and every Voronoi site agent is responsible for monitoring and tracking the target in its cell. Then, a cooperative relay tracking strategy is proposed such that during the tracking process, when a target enters a new Voronoi cell, this event triggers the switching of both tracking agents and communication topology. This is significantly different from the traditional switching topologies. In addition, during the tracking process, the topology and tracking agents switch, which may lead the tracking system to be stable or unstable. The system switches either among consecutive stable subsystems and consecutive unstable subsystems or between stable and unstable subsystems. The objective of this paper is to design a tracking strategy guaranteeing overall successful tracking despite the existence of unstable subsystems. We also address extended discussions on the case where the dynamics of agents are subject to disturbances and the disturbance attenuation level is achieved. Finally, the proposed tracking strategy is verified by a set of simulations.

Trajectory optimization of space maneuver vehicle using a hybrid optimal control solver

In this paper, a constrained space maneuver vehicles trajectory optimization problem is formulated and solved using a new three-layer-hybrid optimal control solver. To decrease the sensitivity of the initial guess and enhance the stability of the algorithm, an initial guess generator based on a specific stochastic algorithm is applied. In addition, an improved gradient-based algorithm is used as the inner solver, which can offer the user more flexibility to control the optimization process. Furthermore, in order to analyze the quality of the solution, the optimality verification conditions are derived. Numerical simulations were carried out by using the proposed hybrid solver and the results indicate that the proposed strategy can have better performance in terms of convergence speed and convergence ability when compared with other typical optimal control solvers. A Monte-Carlo simulation was performed and the results show a robust performance of the proposed algorithm in dispersed conditions.

Solving multi-objective aeroassisted spacecraft trajectory optimization problems using extended NSGA-II

This paper focuses on the application of an extended NSGA-II algorithm to the aeroassisted spacecraft trajectory optimization problems. A new multi-objective spacecraft optimal control model is formulated and parameterized using discretization method. The resulting multi-objective nonlinear programming problem is then solved via the multiobjective evolutionary solver. In order to deal with path constraints that naturally arise in practical trajectory planning problems, the original NSGA-II approach is extended by introducing a new constraint handling strategy. Simulation results are provided to illustrate the effectiveness and feasibility of the enhanced NSGA-II algorithm in dealing with spacecraft trajectory optimization problems.

Modified bidirectional extreme learning machine with Gram–Schmidt orthogonalization method

Abstract Incremental extreme learning machine has been proved to be an efficient and simple universal approximator. However, the network architecture may be very large due to the inefficient nodes which have a tiny effect on reducing the residual error. More to the point, the output weights are not the least square solution. To reduce such inefficient nodes, a method called bidirectional ELM (B-ELM), which analytically calculates the input weights of even nodes, was proposed. By analyzing, B-ELM can be further improved to achieve better performance on compacting structure. This paper proposes the modified B-ELM (MB-ELM), in which the orthogonalization method is involved in B-ELM to orthogonalize the output vectors of hidden nodes and the resulting vectors are taken as the output vectors. MB-ELM can greatly diminish inefficient nodes and obtain a preferable output weight vector which is the least square solution, so that it has better convergence rate and a more compact network architecture. Specifically, it has been proved that in theory, MB-ELM can reduce residual error to zero by adding only two nodes into network. Simulation results verify these conclusions and show that MB-ELM can reach smaller low limit of residual error than other I-ELM methods.

Unified multiobjective optimization scheme for aeroassisted vehicle trajectory planning

In this work, a multiobjective aeroassisted trajectory optimization problem with mission priority constraints is constructed and studied. To effectively embed the priority requirements into the optimization model, a specific transformation technique is applied and the original problem is then transcribed to a single-objective formulation. The resulting single-objective programming model is solved via an evolutionary optimization algorithm. Such a design is unlike most traditional approaches where the nondominated sorting procedure is required to be performed to rank all the objectives. Moreover, in order to enhance the local search ability of the optimization process, a hybrid gradient-based operator is introduced. Simulation results indicate that the proposed design can produce feasible and high-quality flight trajectories. Comparative simulations with other typical methods are also performed, and the results show that the proposed approach can achieve a better performance in terms of satisfying the pres...

An Extreme Learning Machine-Based Community Detection Algorithm in Complex Networks

Community structure, one of the most popular properties in complex networks, has long been a cornerstone in the advance of various scientific branches. Over the past few years, a number of tools have been used in the development of community detection algorithms. In this paper, by means of fusing unsupervised extreme learning machines and the -means clustering techniques, we propose a novel community detection method that surpasses traditional -means approaches in terms of precision and stability while adding very few extra computational costs. Furthermore, results of extensive experiments undertaken on computer-generated networks and real-world datasets illustrate acceptable performances of the introduced algorithm in comparison with other typical community detection algorithms.