A rule-based robot scheduling system for flexible manufacturing cells

In this research we examine a class of flexible manufacturing cells (FMC) containing a robot. The role of the robot is to load parts onto machines, to unload parts from machines, and to transport parts between machines. Since the productivity of an FMC is directly proportional to the level of productive work performed by the robot, the manner in which robots move between machines affects productivity. The problem of finding efficient robot schedules/tours is therefore one of substantial economic significance in the operation of a FMC. Unfortunately, in many practical situations it is difficult to develop efficient robot schedules given the dynamic environments in which they exist. We devise a rule-based system to assist the cell supervisor in making good decisions by dynamically coordinating the available information during the production process. The rule-based system combines an algorithmic procedure to deal with a well-structured environment and a flexible heuristic approach employed to deal with less well-structured environments. Both the algorithmic and heuristic procedures are applied separately, then together, to control the robot's movement in a simulation experiment. We show that there is a predictable tradeoff between the quality of the resulting schedule and the information contents of heuristic used.     

Modeling and optimization of implementation aspects in industrial robot coordination

Cycle time for an assembly line in the automotive industry is a crucial requirement for the overall production rate and business achievements. In a multi-robot station, besides workload distribution, operations sequencing, and robot motions, an important issue is robot coordination. A key aspect, so far neglected, is the robot motion delay due to the implementation of robot coordination scheduling: our first contribution is to model these delays into a mathematical formulation in order to minimize their contribution to the overall cycle time. Two different setup scenarios, based on typical hardware setups, are modeled and optimized. The major contribution is an efficient heuristic algorithm to optimize our suggested model, which provides near optimal solutions in fraction of the time required by a general optimization package. We have also integrated such algorithm in a CAM software for robot offline simulation to be able to run industrial scenarios with real CAD models from the automotive industry. The results show that it is possible to reduce the delays due to the synchronization points and consequently cycle time, while still maintaining the pre-computed nominal time scheduling for robot coordination.     

Hybrid discrete differential evolution algorithm for biobjective cyclic hoist scheduling with reentrance

Cyclic hoist scheduling problems in automated electroplating lines and surface processing shops attract many attentions and interests both from practitioners and researchers. In such systems, parts are transported from a workstation to another by a material handling hoist. The existing literature mainly addressed how to find an optimal cyclic schedule to minimize the cycle time that measures the productivity of the lines. The material handling cost is an important factor that needs to be considered in practice but seldom addressed in the literature. This study focuses on a biobjective cyclic hoist scheduling problem to minimize the cycle time and the material handling cost simultaneously. We consider the reentrant workstations that are usually encountered in real-life lines but inevitably make the part-flow more complicated. The problem is formulated as a biobjective linear programming model with a given hoist move sequence and transformed into finding a set of Pareto optimal hoist move sequences with respect to the bicriteria. To obtain the Pareto optimal or near-optimal front, a hybrid discrete differential evolution (DDE) algorithm is proposed. In this hybrid evolutional algorithm, the population is divided into several subpopulations according to the maximal work-in-process (WIP) level of the system and the sizes of subpopulations are dynamically adjusted to balance the exploration and exploitation of the search. We propose a constructive heuristic to generate initial subpopulations with different WIP levels, hybrid mutation and crossover operators, an evaluation method that can tackle infeasible individuals and a one-to-one greedy tabu selection method. Computational results on both benchmark instances and randomly generated instances show that our proposed hybrid DDE algorithm outperforms the basic DDE algorithm and can solve larger-size instances than the existing ε-constraint method.     

通过网格改进的基于指标的进化算法

设计一种高效的演化多目标优化算法,使其能获得一组同时具有优异的收敛性和多样性的解集是一项很困难的任务。为了能高效求解多目标优化问题,在基于指标的进化算法(IBEA)的基础上：1)引入基于目标空间网格的多样性保持策略,保证算法近似前沿具有优异的分布性;2)引入反向学习机制,同时评估当前解和当前解的反向解,期望能找到一组较优的解从而加快算法收敛。通过6个标准测试函数对改进算法进行测试,其结果表明改进算法可以有效逼近真实Pareto前沿并且分布均匀。     

Multiobjective pseudo‐variable neighborhood descent for a bicriteria parallel machine scheduling problem with setup time

A multiobjective variable neighborhood descent (VND) based heuristic is developed to solve a bicriteria parallel machine scheduling problem. The problem considers two objectives, one related to the makespan and the other to the flow time, where the setup time depends on the sequence, and the machines are identical. The heuristic has a set of neighborhood structures based on swap, remove, and insertion moves. We propose changing the local search inside the VND to a sequential search through the neighborhoods to obtain nondominated points for the Pareto‐front quickly. In the numerical tests, we consider a single‐objective version of the heuristic, comparing the results on 510 benchmark instances to show that it is quite effective. Moreover, new instances are generated in accordance with the literature for the bicriteria problem, showing the ability of the proposed heuristic to return an efficient set of nondominate solutions compared with the well‐known nondominated sorting genetic algorithm II.     

Non-additive multi-objective robot coalition formation

Manifold increase in the complexity of robotic tasks has mandated the use of robotic teams called coalitions that collaborate to perform complex tasks. In this scenario, the problem of allocating tasks to teams of robots (also known as the coalition formation problem) assumes significance. So far, solutions to this NP-hard problem have focused on optimizing a single utility function such as resource utilization or the number of tasks completed. We have modeled the multi-robot coalition formation problem as a multi-objective optimization problem with conflicting objectives. This paper extends our recent work in multi-objective approaches to robot coalition formation, and proposes the application of the Pareto Archived Evolution Strategy (PAES) algorithm to the coalition formation problem, resulting in more efficient solutions. Simulations were carried out to demonstrate the relative diversity in the solution sets generated by PAES as compared to previously studied methods. Experiments also demonstrate the relative scalability of PAES. Finally, three different selection strategies were implemented to choose solutions from the Pareto optimal set. Impact of the selection strategies on the final coalitions formed has been shown using Player/Stage.     

A novel methodology for optimal single mobile robot scheduling using whale optimization algorithm

One of the fundamental requirements for creating an intelligent manufacturing environment is to develop a reliable, efficient and optimally scheduled material transport system. Besides traditional material transport solutions based on conveyor belts, industrial trucks, or automated guided vehicles, nowadays intelligent mobile robots are becoming widely used to satisfy this requirement. In this paper, the authors analyze a single mobile robot scheduling problem in order to find an optimal way to transport raw materials, goods, and parts within an intelligent manufacturing system. The proposed methodology is based on biologically inspired Whale Optimization Algorithm (WOA) and is aimed to find the optimal solution of the nondeterministic polynomial-hard (NP-hard) scheduling problem. The authors propose a novel mathematical model for the problem and give a mathematical formulation for minimization of seven fitness functions (makespan, robot finishing time, transport time, balanced level of robot utilization, robot waiting time, job waiting time, as well as total robot and job waiting time). This newly developed methodology is extensively experimentally tested on 26 benchmark problems through three experimental studies and compared to five meta-heuristic algorithms including genetic algorithm (GA), simulated annealing (SA), generic and chaotic Particle Swarm Optimization algorithm (PSO and cPSO), and hybrid GA–SA algorithm. Furthermore, the data are analyzed by using the Friedman statistical test to prove that results are statistically significant. Finally, generated scheduling plans are tested by Khepera II mobile robot within a laboratory model of the manufacturing environment. The experimental results show that the proposed methodology provides very competitive results compared to the state-of-art optimization algorithms.     

A neural network approach for a robot task sequencing problem

This paper presents a neural network approach with successful implementation for the robot task-sequencing problem. The problem addresses the sequencing of tasks comprising loading and unloading of parts into and from the machines by a material-handling robot. The performance criterion is to minimize a weighted objective of the total robot travel time for a set of tasks and the tardiness of the tasks being sequenced. A three-phased parallel implementation of the neural network algorithm on Thinking Machine's CM-5 parallel computer is also presented which resulted in a dramatic increase in the speed of finding solutions. To evaluate the performance of the neural network approach, a branch-and-bound method and a heuristic procedure have been developed for the problem. The neural network method is shown to give good results and is especially useful for solving large problems on a parallel-computing platform.     

A heuristic algorithm based on Lagrangian relaxation for the closest string problem

The closest string problem that arises in both computational biology and coding theory is to find a string minimizing the maximum Hamming distance from a given set of strings. This study proposes an efficient heuristic algorithm for this NP-hard problem. The key idea is to apply the Lagrangian relaxation technique to the problem formulated as a mixed-integer programming problem. This enables us to decompose the problem into trivial subproblems corresponding to each position of the strings. Furthermore, a feasible solution can be easily obtained from a solution of the relaxation. Based on this, a heuristic algorithm is constructed by combining a Lagrangian multiplier adjustment procedure and a tabu search. Computational experiments will show that the proposed algorithm can find good approximate solutions very fast.     

改进的势场蚁群算法的移动机器人路径规划

针对蚁群算法路径规划初期信息素浓度差异较小,正反馈作用不明显,路径搜索存在着盲目性、收敛速度相对较慢、易陷入局部最优等情况,人工势场算法的势场力可引导机器人快速朝目标位置前进,提出势场蚁群算法,通过栅格法对机器人的工作环境进行建模,利用人工势场中的势场力、势场力启发信息影响系数及蚁群算法中机器人与目标位置的距离构造综合启发信息,并利用蚁群算法的搜索机制在未知环境中寻找一条最优路径。大量的仿真实验表明势场蚁群算法路径规划能找到更优路径和收敛速度更快。     

带机器人制造单元的作业车间调度仿真

针对带有机器人制造单元的作业车间调度优化问题, 在若干加工机器上可以加工具有特定加工工序的若干工件, 并且搬运机器人可以将工件在装卸载站与各加工机器间进行搬运. 在实际生产过程中, 由于不确定性, 特别是带有存货的加工单元, 要求工件的完工时间在一个时间窗内, 而不是一个特定的时间点. 因此针对此情况的作业车间, 考虑到其在求解问题过程中的复杂性和约束性等特点, 研究了在时间窗约束下, 目标值为最小化工件完成时间提前量和延迟量的总权重. 提出了一种将文化基因算法与邻域搜索技术(变邻域下降搜索)相结合的改进元启发式算法, 在求得最优目标值的同时, 可得到最优值的工件加工序列及机器人搬运序列. 通过实验结果表明, 所提出的算法有效且优于传统文化基因算法与遗传算法.     

A non-delayed relax-and-cut algorithm for scheduling problems with parallel machines,due dates and sequence-dependent setup times

Consider the problem of scheduling a set of jobs to be processed exactly once, on any machine of a set of unrelated parallel machines, without preemption. Each job has a due date, weight, and, for each machine, an associated processing time and sequence-dependent setup time. The objective function considered is to minimize the total weighted tardiness of the jobs.This work proposes a non-delayed relax-and-cut algorithm, based on a Lagrangean relaxation of a time indexed formulation of the problem. A Lagrangean heuristic is also developed to obtain approximate solutions.Using the proposed methods, it is possible to obtain optimal solutions within reasonable time for some instances with up to 180 jobs and six machines. For the solutions for which it is not possible to prove optimality, interesting gaps are obtained.     

A stochastic model for the analysis of a two-machine flexible manufacturing cell

This paper presents a stochastic model to determine the performance of a flexible manufacturing cell (FMC) under variable operational conditions, including random machining times, random loading and unloading times, and random pallet transfer times. The FMC under study consists of two machines, pallet handling system, and a loading/unloading robot. After delivering the blanks by the pallet to the cell, the robot loads the first machine followed by the second. Unloading of a part starts with the machine that finishes its part first, followed by the next machine. When the machining of all parts on the pallet is completed, the handling system moves the pallet with finished parts out and brings in a new pallet with blanks. A model with these characteristics turns out to be a Markov chain with a transition matrix of size 5n+3, where n is the number of parts on the pallet. In this paper, we present exact numerical solutions and economic analysis to evaluate FMC systems, to determine optimal pallet capacity and robot speed that minimize total FMC cost per unit of production.     

Symbolically efficient formulations for computational robot dynamics

In 1983, the authors implemented the computer program Algebraic Robot Modeler (ARM) to generate symbolically complete closed-form and recursive dynamic robot models.^1–4 Then, in 1985, we incorporated in ARM heuristic rules for the systematic organization of dynamic robot models to reduce the computational requirements of customized forward and inverse dynamics calculations. We compare the symbolic efficiencies of six robot dynamics formulations for generating closed-form and recursive models. We find that our Lagrange-Christoffel formulation is the most symbolically efficient generator of closed-form dynamic robot models. In our companion paper,³ we resolve the issue of numerical efficiency of customized closed-form and recursive algorithms for computing the forward and inverse dynamics of kinematically and dynamically structured manipulators.     

Heuristic solution for constrained 7-DOF motion planning in 3D scanning application

This paper proposes a heuristic algorithm for near-optimal handling of the redundancy in robot mechanisms, applied to a 3D scanning platform consisting of a 6-DOF articulated robot arm which moves a laser sensor around a workpiece placed on a rotary table. The proposed method handles both static constraints, like avoiding joint limits, kinematic singularities or collisions, modelled using a configuration map viewed as a grayscale image, and dynamic constraints like velocity and acceleration. The algorithm is compared with Dijkstra algorithm, which gives the optimal solution, but is very slow, and with two other heuristics which are very fast, but give suboptimal solutions.     

针对移动云计算任务迁移的快速高效调度算法

计算量较大的应用程序由于需要大量的能耗,因此在电池容量有限的移动设备上运行时十分受限。云计算迁移技术是保证此类应用程序在资源有限的设备上运行的主流方法。针对无线网络中应用程序任务图的调度和迁移问题,提出了一种快速高效的启发式算法。该算法将能够迁移到云端的任务都安排在云端完成这种策略作为初始解,通过逐次计算可迁移任务在移动端运行的能耗节省量,依次将节省量最大的任务迁移到移动端,并依据任务间的通讯时间及时更新各个任务的能耗节省量。为了寻找全局最优解,构造了适用于此问题的禁忌搜索算法,给出了相应的编码方法、禁忌表、邻域解以及算法终止准则。构造的禁忌搜索算法以提出的启发式解为初始解进行全局搜索,并实现对启发解的进一步优化。通过 实验 将所提方法与无迁移、随机迁移、饱和迁移3类算法进行对比,结果表明提出的启发式算法能够快速有效地给出能耗更小的解。例如,在宽度为10的任务图上,当深度为8时,无迁移、随机迁移与饱和迁移的能耗分别为5461、3357和2271能量单位,而给出的启发解对应的能耗仅为2111。在此基础上禁忌搜索算法又将其能耗降低到1942, 这进一步说明了提出的启发式算法能够产生高质量的近似解。     

Evolutionary Optimization of File Assignment for a Large-Scale Video-on-Demand System

We present a genetic algorithm for tackling a file assignment problem for a large-scale video-on-demand system. The file assignment problem is to find the optimal replication and allocation of movie files to disks so that the request blocking probability is minimized subject to capacity constraints. We adopt a divide-and-conquer strategy, where the entire solution space of file assignments is divided into subspaces. Each subspace is an exclusive set of solutions sharing a common file replication instance. This allows us to utilize a greedy file allocation method for finding a good-quality heuristic solution within each subspace. We further design two performance indices to measure the quality of the heuristic solution on 1.) its assignment of multicopy movies and 2.) its assignment of single-copy movies. We demonstrate that these techniques, together with ad hoc population handling methods, enable genetic algorithms to operate in a significantly reduced search space and achieve good-quality file assignments in a computationally efficient way.     

Minimizing makespan subject to minimum flowtime on two identical parallel machines

We consider the problem of scheduling jobs on two parallel identical machines where an optimal schedule is defined as one that gives the smallest makespan (the completion time of the last job) among the set of schedules with optimal total flowtime (the sum of the completion times of all jobs). We propose an algorithm to determine optimal schedules for the problem, and describe a modified multifit algorithm to find an approximate solution to the problem in polynomial computational time. Results of a computational study to compare the performance of the proposed algorithms with a known heuristic shows that the proposed heuristic and optimization algorithms are quite effective and efficient in solving the problem.Scope and purposeMultiple objective optimization problems are quite common in practice. However, while solving scheduling problems, optimization algorithms often consider only a single objective function. Consideration of multiple objectives makes even the simplest multi-machine scheduling problems NP-hard. Therefore, enumerative optimization techniques and heuristic solution procedures are required to solve multi-objective scheduling problems. This paper illustrates the development of an optimization algorithm and polynomially bounded heuristic solution procedures for the scheduling jobs on two identical parallel machines to hierarchically minimize the makespan subject to the optimality of the total flowtime.     

基于强化学习的煤矸石分拣机械臂智能控制算法研究

针对传统煤矸石分拣机械臂控制算法如抓取函数法、基于费拉里法的动态目标抓取算法等依赖于精确的环境模型、且控制过程缺乏自适应性,传统深度确定性策略梯度(DDPG)等智能控制算法存在输出动作过大及稀疏奖励容易被淹没等问题,对传统DDPG算法中的神经网络结构和奖励函数进行了改进,提出了一种适合处理六自由度煤矸石分拣机械臂的基于强化学习的改进DDPG算法。煤矸石进入机械臂工作空间后,改进DDPG算法可根据相应传感器返回的煤矸石位置及机械臂状态进行决策,并向相应运动控制器输出一组关节角状态控制量,根据煤矸石位置及关节角状态控制量控制机械臂运动,使机械臂运动到煤矸石附近,实现煤矸石分拣。仿真实验结果表明:改进DDPG算法相较于传统DDPG算法具有无模型通用性强及在与环境交互中可自适应学习抓取姿态的优势,可率先收敛于探索过程中所遇的最大奖励值,利用改进DDPG算法控制的机械臂所学策略泛化性更好、输出的关节角状态控制量更小、煤矸石分拣效率更高。     

Scheduling a single mobile robot for part-feeding tasks of production lines

This study deals with the problem of sequencing feeding tasks of a single mobile robot which is able to provide parts for feeders of machines on production lines. The mobile robot has to be scheduled in order to stoppage from lack of parts in the production line. A method based on the characteristics of feeders and inspired by the ( \(s,Q\) ) inventory system, is thus applied to define time windows for the feeding tasks of the robot. The capacity of the robot is also taken into consideration. The performance criterion is to minimize total traveling time of the robot for a given planning horizon. A genetic algorithm-based heuristics is presented which results in a significant increase in the speed of finding near-optimal solutions. To evaluate the performance of the genetic algorithm-based heuristic, a mixed-integer programming model has been developed for the problem. A case study is implemented at an impeller production line in a real factory and computational experiments are also conducted to demonstrate the effectiveness of the proposed approach.