Multi-objective optimization of inverse planning for accurate radiotherapy

The multi-objective optimization of inverse planning based on the Pareto solution set, according to the multi-objective character of inverse planning in accurate radiotherapy, was studied in this paper. Firstly, the clinical requirements of a treatment plan were transformed into a multi-objective optimization problem with multiple constraints. Then, the fast and elitist multi-objective Non-dominated Sorting Genetic Algorithm (NSGA-Ⅱ) was introduced to optimize the problem. A clinical example was tested using this method. The results show that an obtained set of non-dominated solutions were uniformly distributed and the corresponding dose distribution of each solution not only approached the expected dose distribution, but also met the dose-volume constraints. It was indicated that the clinical requirements were better satisfied using the method and the planner could select the optimal treatment plan from the non-dominated solution set.     

Multi-objective optimization of inverse planning for accurate radiotherapy

The multi-objective optimization of inverse planning based on the Pareto solution set, according to the multi-objective character of inverse planning in accurate radiotherapy, was studied in this paper. Firstly,the clinical requirements of a treatment plan were transformed into a multi-objective optimization problem with multiple constraints. Then, the fast and elitist multi-objective Non-dominated Sorting Genetic Algorithm (NSGA-Ⅱ)was introduced to optimize the problem. A clinical example was tested using this method. The results show that an obtained set of non-dominated solutions were uniformly distributed and the corresponding dose distribution of each solution not only approached the expected dose distribution, but also met the dosevolume constraints. It was indicated that the clinical requirements were better satisfied using the method and the planner could select the optimal treatment plan from the non-dominated solution set.     

A novel approach for multiple mobile objects path planning: Parametrization method and conflict resolution strategy

We present a new approach containing two steps to determine conflict-free paths for mobile objects in two and three dimensions with moving obstacles. Firstly, the shortest path of each object is set as goal function which is subject to collision-avoidance criterion, path smoothness, and velocity and acceleration constraints. This problem is formulated as calculus of variation problem (CVP). Using parametrization method, CVP is converted to time-varying nonlinear programming problems (TNLPP) and then resolved. Secondly, move sequence of object is assigned by priority scheme; conflicts are resolved by multilevel conflict resolution strategy. Approach efficiency is confirmed by numerical examples.     

改进蚁群算法的局部信息动态路径规划

针对传统蚁群算法收敛速度慢、对动态路径变化适应性低的局限性,提出了一种基于局部信息获取策略的动态改进型蚁群算法。该算法利用局部信息获取策略,进行最优局部目标点的获取,然后调用改进蚁群算法获取局部区域内的最优路径,再重复循环获取新的最优局部目标点,直到找到全局目标点。与此同时,将提出的改进型蚁群算法应用于动态路径规划中的路径寻优与避障,仿真结果表明：提出的算法在具有与传统蚁群算法相当的路径优化效果的同时,能够有效适应障碍变化、大大提高了路径规划的收敛速度。     

超-超引射器多目标优化设计

推导出了超-超引射器性能计算和优化设计模型,借助Pareto优胜、Pareto最优解和Pareto前端等概念,采用基于多目标进化/分解算法（MOEA/D）的多目标优化方法,计算得到超-超引射器多目标优化问题的Pareto前端,解决了超-超引射器多目标优化设计问题,并与常规参数分析方法进行了比较。结果表明:超-超引射器性能影响参数相互关系复杂,增压比和引射系数作为引射器主要性能参数相互冲突,通过常规分析难以得到较清晰的设计准则,利用多目标优化设计方法可有效地辅助多属性决策和系统优化设计。     

Region coverage-aware path planning for unmanned aerial vehicles: A systematic review

The use of unmanned aerial vehicles (UAVs) to carry out remote aerial surveys has become prominent in recent years. The UAV-based survey faces several operational issues, such as complicated terrain, limited UAV resources, obstacles, sensor limitations, and other environmental factors. In addition, the coverage plan includes numerous objectives, such as lowering path length, maximizing coverage, and limiting survey time, necessitating multi-objective optimization. UAVs require effective coverage path planning (CPP) to generate the ideal route. It involves determining the path which encompasses every point inside the required region under different constraints. The process automates the process of route determination for autonomous operation by considering various environmental constraints. In this paper, we explore and analyze the existing research on the various techniques used in coverage route planning for UAVs. It provides an overview of the current state-of-the-art CPP methods for UAVs. The study discusses the key challenges and requirements of CPP for UAVs and presents various approaches proposed in the literature to tackle these challenges. We explore a variety of geometric flight patterns for the area of interest having UAV deployment. It also features multi-UAV and multi-region coverage strategies, providing a new dimension to UAV-based operations. The energy consumption of UAVs during CPP is an essential factor, as it influences their flight length and mission duration. The design of the CPP algorithm is determined by the unique requirements of the UAV application, such as the size and form of the region to be mapped, the existence of obstacles, and the desired coverage resolution. Path planning strategies in a three-dimensional environment and dynamic coverage are also included in the study. Moreover, we compare the existing strategies using different performance metrics to evaluate the success of covering missions. Finally, the problems and unresolved concerns related to UAV coverage path planning are examined to provide valuable insights to the readers.     

基于激光雷达的移动机器人动态障碍检测

动态环境下运动物体的检测是移动机器人研究的难点问题之一。以未知环境为研究背景,提出了一种基于激光雷达的自主动态障碍检测方法。通过k-近邻方法对激光雷达测距数据进行了空间障碍聚类,在此基础上分析了聚类障碍的特征参数,例如区域、质心,利用聚类障碍数据的时间关联性分析并确定了障碍的类型,并采用模糊地图匹配的策略实现了地图匹配和更新。在所研制的移动机器人上进行了实验,实验结果验证了方法的有效性。     

A Multi-Objective Multi-Label Feature Selection Algorithm Based on Shapley Value

Multi-label learning is dedicated to learning functions so that each sample is labeled with a true label set. With the increase of data knowledge, the feature dimensionality is increasing. However, high-dimensional information may contain noisy data, making the process of multi-label learning difficult. Feature selection is a technical approach that can effectively reduce the data dimension. In the study of feature selection, the multi-objective optimization algorithm has shown an excellent global optimization performance. The Pareto relationship can handle contradictory objectives in the multi-objective problem well. Therefore, a Shapley value-fused feature selection algorithm for multi-label learning (SHAPFS-ML) is proposed. The method takes multi-label criteria as the optimization objectives and the proposed crossover and mutation operators based on Shapley value are conducive to identifying relevant, redundant and irrelevant features. The comparison of experimental results on real-world datasets reveals that SHAPFS-ML is an effective feature selection method for multi-label classification, which can reduce the classification algorithm’s computational complexity and improve the classification accuracy.     

基于Pareto优化理论的多目标超椭梯度线圈设计

磁共振系统梯度线圈设计是一个多目标优化问题,在设计时需要综合考虑能耗、磁场能、线性度等设计要求.这些设计要求通常难以同时获得极小解,因此在设计梯度线圈时需要权衡线圈的各方面的设计需求.本文基于柱面可展性和流函数设计方法,结合Pareto优化方法实现了在超椭圆柱设计表面上梯度线圈的多目标设计.分别分析了磁场能、能耗目标对梯度线圈线性度、线圈构型的影响;并在Pareto解空间中分析各目标的相互变化关系,通过数值算例验证了该方法在超椭梯度线圈设计时的有效性与灵活性.优化结果显示,在满足线性度误差小于5%,能耗与磁场能分别小于用户设定值的设计约束下,梯度线圈的多目标设计存在多个局部优化解.该方法可以直观地比较相同目标函数值的情况下各单目标的具体表现,有利于实现不同的设计要求下梯度线圈的最终定型设计.     

基于遗传算法的输电线路无人机巡检路径规划

针对无人机有效、安全巡检输电线路的路径问题,提出了一种基于遗传算法的输电线路无人机巡检路径规划方法,采用极坐标编码方式对无人机巡检路径构造染色体。结合实际情况中的无人机巡检各种约束问题,设计了适合于无人机巡检路径规划的遗传算子。实验结果证明算法能综合考虑各种因素,提高了全局寻优能力,是解决实际输电线路无人机巡检路径规划问题的较好办法。     

Multi-objective optimal design and experimental validation of tracking control of a rotating flexible beam

This paper investigates the tracking control problem for a rotating flexible beam. A proportional-integral-derivative control is designed to meet multiple objectives including overshoot, peak time, tracking error of the rigid movement of the rotating base, and the defectional angle of the flexible beam. A multi-objective optimization problem is then formulated for the control design and is solved with the cell mapping method. Numerical simulations and experiments are carried out to demonstrate the effect of the different control gains in the Pareto set, and to study the difference between a linear and a nonlinear model of the flexible beam.     

目标函数设置对放疗逆向计划多目标优化过程的影响

目前放疗逆向计划中常用的目标函数有两种:基于剂量分布的目标函数和基于剂量．体积直方图（DVH）的目标函数。实际系统都基于单目标优化算法进行,不考虑逆向计划的多目标性。在Pareto多目标优化理论的基础上,研究逆向计划的目标函数设置问题,比较基于剂量分布的目标函数和基于DVH的目标函数对多目标优化过程的影响,包括优化时间、收敛性和存在的问题等,为逆向计划过程中多目标优化目标函数的设置提供依据。There are two kinds of objective functions in radiotherapy inverse planning: dose distribution-based and Dose-Volume Histogram （DVH）-based functions. The treatment planning in our days is still a trial and error process because the multi-objective problem is solved by transforming it into a single objective problem using a specific set of weights for each object. This work investigates the problem of objective function setting based on Pareto multi-optimization theory, and compares the effect on multi-objective inverse planning of those two kinds of objective functions including calculation time, converge speed, etc. The basis of objective function setting on inverse planning is discussed.     

Piezoelectric transducer design via multiobjective optimization

The design of piezoelectric transducers is usually based on single-objective optimization only. In most practical applications of piezoelectric transducers, however, there exist multiple design objectives that often are contradictory to each other by their very nature. It is impossible to find a solution at which each objective function gets its optimal value simultaneously. Our design approach is to first find a set of Pareto-optimal solutions, which can be considered to be best compromises among multiple design objectives. Among these Pareto-optimal solutions, the designer can then select the one solution which he considers to be the best one. In this paper we investigate the optimal design of a Langevin transducer. The design problem is formulated mathematically as a constrained multiobjective optimization problem. The maximum vibration amplitude and the minimum electrical input power are considered as optimization objectives. Design variables involve continuous variables (dimensions of the transducer) and discrete variables (the number of piezoelectric rings and material types). In order to formulate the optimization problem, the behavior of piezoelectric transducers is modeled using the transfer matrix method based on analytical models. Multiobjective evolutionary algorithms are applied in the optimization process and a set of Pareto-optimal designs is calculated. The optimized results are analyzed and the preferred design is determined.     

叶轮机械叶栅多目标多学科设计优化算法

提出了基于多目标寻优的叶轮机械叶栅多学科设计优化算法,方法包括:采用并行多目标差分进化算法作为优化求解器来搜寻叶栅多学科设计优化问题的Paxeto解集,采用非均匀B样条方法对叶片型面进行参数化处理,通过求解Reynolds-Avergaed Navier-Stokes方程评估叶片的气动性能,耦合气动计算得到的叶片表面压力,应用有限元分析方法预测叶片的强度性能.为证明本文方法的实用性,选择叶片的等熵效率和叶片应力为目标函数,完成了NASA Rotor 37转子叶栅的多学科设计优化,结果表明本文提出的多学科设计优化算法具有良好的优化性能.     

Genetic algorithms for the QoS based multicast routing and wavelength allocation problem in WDM network

To tackle the QoS based multicast routing and wavelength allocation problem (MRWA), three multi-objective genetic algorithms are proposed, which are based on the ideas of Non-dominated Sorting, Strength Pareto and Decomposition, respectively. The chromosome coding scheme, crossover and mutation operators are redefined. To ensure the generated offspring being a connected light-tree, a light-path repair process and a loop eliminating process are designed. The proposed algorithms were evaluated on a set of different scale test problems and compared with the recently proposed GA based multi-objective optimization algorithm for this problem. The experimental results reveal very encouraging results in terms of the solution quality.     

A new multi-objective approach to finite element model updating

The single objective function (SOF) has been employed for the optimization process in the conventional finite element (FE) model updating. The SOF balances the residual of multiple properties (e.g., modal properties) using weighting factors, but the weighting factors are hard to determine before the run of model updating. Therefore, the trial-and-error strategy is taken to find the most preferred model among alternative updated models resulted from varying weighting factors. In this study, a new approach to the FE model updating using the multi-objective function (MOF) is proposed to get the most preferred model in a single run of updating without trial-and-error. For the optimization using the MOF, non-dominated sorting genetic algorithm-II (NSGA-II) is employed to find the Pareto optimal front. The bend angle related to the trade-off relationship of objective functions is used to select the most preferred model among the solutions on the Pareto optimal front. To validate the proposed approach, a highway bridge is selected as a test-bed and the modal properties of the bridge are obtained from the ambient vibration test. The initial FE model of the bridge is built using SAP2000. The model is updated using the identified modal properties by the SOF approach with varying the weighting factors and the proposed MOF approach. The most preferred model is selected using the bend angle of the Pareto optimal front, and compared with the results from the SOF approach using varying the weighting factors. The comparison shows that the proposed MOF approach is superior to the SOF approach using varying the weighting factors in getting smaller objective function values, estimating better updated parameters, and taking less computational time.     

Robust optimization of nonlinear impulsive rendezvous with uncertainty

The optimal rendezvous trajectory designs in many current research efforts do not incorporate the practical uncertainties into the closed loop of the design.A robust optimization design method for a nonlinear rendezvous trajectory with uncertainty is proposed in this paper.One performance index related to the variances of the terminal state error is termed the robustness performance index,and a two-objective optimization model(including the minimum characteristic velocity and the minimum robustness performance index)is formulated on the basis of the Lambert algorithm.A multi-objective,non-dominated sorting genetic algorithm is employed to obtain the Pareto optimal solution set.It is shown that the proposed approach can be used to quickly obtain several inherent principles of the rendezvous trajectory by taking practical errors into account.Furthermore,this approach can identify the most preferable design space in which a specific solution for the actual application of the rendezvous control should be chosen.     

Multi-objective control optimization for semi-active vehicle suspensions

In this paper we demonstrate a method for determining the optimality of control algorithms based on multiple performance objectives. While the approach is applicable to a broad range of dynamic systems, this paper focuses on the control of semi-active vehicle suspensions. The two performance objectives considered are ride quality, as measured by absorbed power, and thermal performance, as measured by power dissipated in the suspension damper. A multi-objective genetic algorithm (MOGA) is used to establish the limits of controller performance. To facilitate convergence, the MOGA is initialized with popular algorithms such as skyhook control, feedback linearization, and sliding mode control. The MOGA creates a Pareto frontier of solutions, providing a benchmark for assessing the performance of other controllers in terms of both objectives. Furthermore, the MOGA provides insight into the remaining achievable gains in performance.     

A restricted-direction target search approach based on coupled routing and optical sensor tasking optimization

A key issue in UAVs (Unmanned Air Vehicles) and UAV-mounted sensor control is the target search problem: finding targets in minimum time. In this paper, we proposed a restricted-direction target search approach based on coupled routing and optical sensor tasking optimization. In this method, we consider a single UAV, which is equipped with two optical sensors to view a limited large region of the dynamic environment. The UAV moves in the dynamic environment, searches for targets of interest, and is capable of avoiding obstacles and threats immediately. The paths are obtained considering actual maneuverability limitations of the UAV and are evaluated according to optimization of the optical sensor tasks for the duration of the path. Series of comparative experimental results demonstrate that this algorithm makes effective use of the coupled method of optimization and performs significantly better than previously proposed approaches.     

Optimization of the ionization time of an atom with tailored laser pulses: a theoretical study

How fast can a laser pulse ionize an atom? We address this question by considering pulses that carry a fixed time-integrated energy per-area, and finding those that achieve the double requirement of maximizing the ionization that they induce, while having the shortest duration. We formulate this double-objective quantum optimal control problem by making use of the Pareto approach to multi-objective optimization, and the differential evolution genetic algorithm. The goal is to find out how a precise time-profiling of ultra-fast, large-bandwidth pulses may speed up the ionization process. We work on a simple one-dimensional model of hydrogen-like atoms (the Pöschl-Teller potential) that allows to tune the number of bound states that play a role in the ionization dynamics. We show how the detailed shape of the pulse accelerates the ionization, and how the presence or absence of bound states influences the velocity of the process.