基于强化学习的遗传算法求解一种新的钻削路径优化问题

孔加工是机械制造的基本工序之一.针对数控机床的刀具路径优化问题,提出一种新颖的孔加工刀具路径优化模型-----带可决策孔的孔加工多刀具路径优化问题(MTdDPO).在该模型中,工件上的孔分为两类:固定孔和可决策孔.MTdDPO的目标是通过判断可决策孔的路径归属和路径内各孔的加工顺序来实现加工路径长度的最小化.为实现MTdDPO的优化,提出基于强化学习的分段遗传算法(RLSGA).在RLSGA中,种群被视为智能体,智能体的状态是种群的多样性系数,3种不同的分段交叉算子是智能体的动作,智能体的奖励与种群的适应度值和多样性系数的变化有关.针对MTdDPO,新建5个基准测试问题,并在测试问题上将RLSGA与其他4个算法进行对比.结果表明,RLSGA的表现明显优于其他算法,RLSGA能够有效地解决MTdDPO问题.     

基于蚁群优化的点胶控制系统路径规划

点胶是电子封装中的一个重要环节。为了改进传统点胶工艺中胶枪移动轨迹通常按X/Y方向进行优化导致加工效率低下的缺点,引入了基于蚁群优化的路径优化算法。大量的仿真结果表明该方法可以有效缩短胶枪的加工路径长度进而提高生产力,尤其当待加工点接近随机分布的时候,本优化算法的加工路径比传统X/Y方向优化所产生的路径短50%左右。     

基于Dubins曲线和改进A*算法的AUV路径规划方法

将Dubins曲线和具有角度约束的改进A*搜索算法结合应用于路径规划中,能解决路径长度最短和安全性的问题。这样规划出来的路径由两段满足AUV最小转弯半径的圆弧和一段同时与两弧相切的直线构成。圆弧段由产生Dubins路径的方法产生,直线段由改进A*搜索算法扩展产生。首先通过判断Dubins路径存在条件,解算Dubins曲线参数,从而确定此路径中两圆弧的起始点、终止点坐标。再通过这些圆弧坐标可得到直线与圆弧的切入点、切出点,此两点就是改进A*搜索算法扩展路径的起始点和终止点。以Matlab为工具进行仿真实验,验证了此方法能产生规避障碍物的可行的最短路径。     

数控加工中路径规划与速度插补综述

数控技术在现代制造工业中被广泛使用,相关研究一直为学界和业界共同关注。数控技术的传 统流程主要包含刀具路径规划和进给速度插补。为实现高速高精加工,人们通常将路径规划与速度插补中的若 干问题转换成数理优化模型,针对工程应用问题的复杂性,采用分步迭代优化的思路进行求解,但所得的结果 往往只是局部最优解。其次,路径规划与速度插补都是为了加工一个工件曲面,分两步进行处理虽然简化了计 算,但也导致不能进行整体优化。因此,为了更好地开展路径规划与速度插补一体化设计与全局最优求解的研 究,系统性地了解并学习已有的代表性工作是十分有必要的。所以将逐次介绍数控加工中刀具路径规划与速度 插补的相关方法与技术进展,包括基于端铣的加工路径规划;刀轴方向优化;G 代码加工以及拐角过渡;参数 曲线路径的进给速度规划等国内外相关研究以及最新提出的一些新型加工优化方法。     

基于改进粒子群三次Bezier曲线优化的路径规划

为了实现在障碍环境空间下移动机器人的平滑最优路径规划,提出了一种利用Bezier曲线描述路径与改进粒子群优化算法相结合的路径规划方法。借助三次Bezier曲线描述路径,可以将路径规划问题转换为生成Bezier曲线有限个点的位置优化问题,通过改进的具有指数变化的认知因子的粒子群优化算法进行最优路径搜索。仿真实验表明,该算法可以有效地进行平滑的无碰撞路径规划,并具有较强的跳出局部最优的能力。     

改进蚁群和鸽群算法的机器人路径规划

针对复杂环境下移动机器人路径规划中存在的迭代速度慢和路径欠优等问题,提出将全局与局部规划算法相结合的路径规划方法。首先,利用同步双向A^*算法对蚁群算法的信息素进行优化,并对蚁群算法的转移概率和信息素更新机制进行改进,从而使算法的全局寻优速度更快,缩短移动机器人的路径长度;进一步地,将静态路径用于鸽群算法的初始化;然后,利用改进的鸽群算法对移动机器人进行了局部路径规划,通过引入模拟退火准则的方法解决局部最优问题,利用对数S型传递函数对鸽群数量的步长进行优化,从而能更好地避免与动态障碍物的碰撞。最后,利用B样条曲线对路径进行平滑化和重规划。仿真结果表明,该方法在全局静态和局部动态阶段均能生成路径长度短、评价值低的平滑路径,且收敛速度快,适合移动机器人在动态复杂环境中的穿行。     

点云化自由曲面上闭合加工路径数学建模

为了解决点云化曲面上闭合加工路径识别问题,对扫描仪获得的大量点云数据先进行网格化约减,提取含有加工路径点云的轮廓网格,再根据网格中的点云估算出网格的特征点,特征点信息包括三维空间坐标和法向量,将特征点有序化,最后用贝塞尔曲线簇建模。最终加工路线由一条封闭的光滑的自由曲线表示,解决了闭合加工路径的识别问题。     

改进蚁群和鸽群算法的机器人路径规划

针对复杂环境下移动机器人路径规划中存在的迭代速度慢和路径欠优等问题，提出将全局与局部规划算法相结合的路径规划方法。首先，利用同步双向A^*算法对蚁群算法的信息素进行优化，并对蚁群算法的转移概率和信息素更新机制进行改进，从而使算法的全局寻优速度更快，缩短移动机器人的路径长度；进一步地，将静态路径用于鸽群算法的初始化；然后，利用改进的鸽群算法对移动机器人进行了局部路径规划，通过引入模拟退火准则的方法解决局部最优问题，利用对数S型传递函数对鸽群数量的步长进行优化，从而能更好地避免与动态障碍物的碰撞。最后，利用B样条曲线对路径进行平滑化和重规划。仿真结果表明，该方法在全局静态和局部动态阶段均能生成路径长度短、评价值低的平滑路径，且收敛速度快，适合移动机器人在动态复杂环境中的穿行。     

四点三线遗传算法求解旅行商问题

为解决NP完全的旅行商问题,提出一种四点三线遗传算法。该算法特色在两阶段策略,第一阶段是变异算子优化,将汉密尔顿环中所有大于两点的内部路径倒置,并用新极值代替原极值。第二阶段是四点三线优化,将汉密尔顿环分为n个四点三线局部路径并将每个局部路径转化为最优局部路径,将所有局部路径长度求和除以1/3。交叉算子结束后,如子代含有重复位点,将未交叉部分重复位点与交叉部分重复位点对应的父代等位点交换。通过将该算法与传统遗传算法及只进行第一步优化的遗传算法进行比较,采用TSPLIB数据库实例数据,证明该算法有更高的执行效率,有更强的收敛性,适合寻找最短TSP路径。     

测量点数据等残留高度刀具路径规划

在介绍计算几何模型的基础上,提出一种刀具路径算法.首先分别以刀具半径值和残留高度值为等距距离,通过等距计算求出等距点集和残留高度点集;然后以前一行刀具路径为中心构建刀具包络面,并求出刀具包络面与残留高度点集的交点,即过渡点集;最后通过求以过渡点集为中心的刀具包络面与等距点集的交点,得出相邻行刀具轨迹,依次递推,求出所有的刀具路径行.针对计算过程中可能出现的欠切问题,给出了应用密切圆追踪的边界处理方法.通过实例验证了该算法的可行性.与等间距刀具路径生成方法进行比较表明,应用文中算法解决针对测量点数据的数控加工刀具路径生成问题,可缩短刀具路径长度,提高加工效率.     

An optimal approach to multiple tool selection and their numerical control path generation for aggressive rough machining of pockets with free-form boundaries

To machine pockets, especially ones with closed free-form boundary curves, roughing is crucial to part productivity, for this operation alone could take more than 60% of the total machining time. At present, there is a high demand from industry for a new machining technique that can efficiently cut pockets. Aggressive rough machining, in which the largest possible cutters are always employed and are fully immersed in workpieces, can be a solution. Although aggressive roughing is by far the most efficient machining strategy, compared to prior pocketing methods, no computer numerical control (CNC) programming technique has been developed to support it, resulting in few applications in machine shops. To address this urgent industrial need, based on the medial axis transform of a pocket, this work proposes an optimal approach to multiple tool selection and their numerical control (NC) path generation for aggressive roughing of the pocket. First, the NC paths of a specific tool are quickly generated using the pocket’s medial axis transform. Thanks to the unique characteristic of the medial axis transform, the paths can ensure the tool the largest accessible space for pocketing. At the same time, they can guarantee the tool to be free of gouging and interference. Then, an optimization model of selecting multiple cutters and generating their NC paths is built in order to achieve the highest efficiency of the aggressive rough machining. To demonstrate the advantages of this innovative approach, two examples are rendered, and their results are compared to those obtained by the existing methods. This approach can be directly implemented into current CAD/CAM software to promote aggressive rough machining of pockets in industry.     

Pocketing toolpath computation using an optimization method

This paper proposes a new method of pocketing toolpath computation based on an optimization problem with constraints. Generally, the calculated toolpath has to minimize the machining time and respect a maximal effort on the tool during machining. Using this point of view, the toolpath can be considered as the result of an optimization in which the objective is to minimize the travel time and the constraints are to check the forces applied to the tool. Thus a method based on this account and using an optimization algorithm is proposed to compute toolpaths for pocket milling. After a review of pocketing toolpath computation methods, the framework of the optimization problem is defined. A modeling of the problem is then proposed and a solving method is presented. Finally, applications and experiments on machine tools are studied to illustrate the advantages of this method.     

改进GSA算法在协同制造任务分配中的应用

针对网络化协同制造中的任务分配问题,建立了以制造任务完成时间、完成成本、产品工艺质量为目标的多目标优化模型,提出了模型求解的改进遗传模拟退火（Genetic Simulated Annealing,GSA）算法。建立了协同制造任务分配的层次结构模型,应用模糊层次分析法分析了时间、成本和工艺质量等因素在协同制造任务分配过程中的相对重要性。设计了优化模型求解的改进遗传模拟退火算法,并结合具体实例验证了算法的有效性和优越性。     

A mapping-based approach to eliminating self-intersection of offset paths on mesh surfaces for CNC machining

Geometrically, a tool path can be generated by successively offsetting its adjacent path on the surface with a given path interval, which preferably starts from one of the surface boundaries or a primary curve. The key issues involved in offset path planning are the generation of raw offset paths and the elimination of the self-intersection of raw offset paths. Most researches available in this area are focused on how to generate the raw offset paths, however, the latter, especially how to eliminate the self-intersection of the offset paths on mesh surfaces, has not been sufficiently addressed. In this paper, a mapping-based approach to eliminating the self-intersection of offset paths is proposed for the CNC machining of mesh surfaces. The method first flattens the mesh surface onto a predefined plane by using a mesh mapping technique, and then taking the mapping as a guide, the offset paths are also naturally mapped onto the plane, from which those invalid self-intersection loops can be effectively identified and eliminated. To handle the issue of self-intersection for all types of offset path, a notion of local loop is introduced to detect and eliminate the invalid self-intersection loops. After that the planar paths are inversely mapped into the physical space and the final tool paths used for the machining of mesh surface are obtained. Meanwhile, in order to improve the kinematic and dynamic performance of the machine tool when machining along the generated offset paths, a method for rounding the sharp corners of tool paths, which result from the process of eliminating the self-intersection of raw offset paths, is also preliminarily investigated. Finally, the proposed method is validated by the results of simulations and machining experiments.     

AGV作业调度模型及改进的DE算法研究

AGV作业调度问题的求解结果对AS/RS的运行效率具有重要影响。通过必要的简化,建立了AGV作业调度问题的静态优化模型。可知静态AGV作业调度问题实质是一种带约束的多重TSP问题,属于典型的NP完全问题,目前还不存在可在多项式时间内求解的确定算法。提出了一种改进的差分演化算法用于求解该问题。为了适应AGV作业调度问题的特点,新算法设计了新的两段编码方法,对多个DE算子进行了改造。还提出了基于生存时间的种群多样性增强机制,用于增强算法的搜索能力,避免陷入局部最优。仿真实验显示,该算法可以有效提高AGV作业调度的效率,验证了相关改进机制的有效性。     

Quart-parametric interpolations for intersecting paths

The intersecting path is an important tool path generation method. This paper proposes an approach for the quart-parametric interpolation of intersecting paths. The objective of our approach is that the intersecting paths for surface machining can be directly interpolated within the computer numerical control (CNC) system. This enables the CNC interpolator to process the intersecting paths without geometric approximation as in existing approaches and take into consideration any specific feedrate profiles and further machining dynamical issues along the path.The interpolation of the intersection of two general parametric surfaces is transferred into interpolation of its projection curves and the time trajectories of four parameters along the intersecting curves are obtained. Our strategy is to carry out the quart-parametric interpolation based on the projection interpolation. The feedrate control method is developed, and then the interpolation algorithms for two projection curves are proposed. An error reduction scheme is presented to alleviate point deviation from the drive parametric surface. Simulations of quart-parametric interpolation have been carried out to verify the effectiveness of the proposed algorithm.     

Software prototype for solving multi-objective machining optimization problems: Application in non-conventional machining processes

For an effective and efficient application of machining processes it is often necessary to consider more than one machining performance characteristics for the selection of optimal machining parameters. This implies the need to formulate and solve multi-objective optimization problems. In recent years, there has been an increasing trend of using meta-heuristic algorithms for solving multi-objective machining optimization problems. Although having the ability to efficiently handle highly non-linear, multi-dimensional and multi-modal optimization problems, meta-heuristic algorithms are plagued by numerous limitations as a consequence of their stochastic nature. To overcome some of these limitations in the machining optimization domain, a software prototype for solving multi-objective machining optimization problems was developed. The core of the developed software prototype is an algorithm based on exhaustive iterative search which guarantees the optimality of a determined solution in a given discrete search space. This approach is justified by a continual increase in computing power and memory size in recent years. To analyze the developed software prototype applicability and performance, four case studies dealing with multi-objective optimization problems of non-conventional machining processes were considered. Case studies are selected to cover different formulations of multi-objective optimization problems: optimization of one objective function while all the other are converted into constraints, optimization of a utility function which combines all objective functions and determination of a set of Pareto optimal solutions. In each case study optimization solutions that had been determined by past researchers using meta-heuristic algorithms were improved by using the developed software prototype.     

面轨道特征的矩形排样与加工路径优化方法研究

为了解决实际生产中遇到的一种带有面轨道特征的矩形排样问题,重点研究了自 适应遗传算法和图论相结合的优化方法,极大提高了切削加工效率。该方法将路径优化问题转 化为一个考察无向图连通性问题,并利用遗传算法在解空间中进行全局搜索,以寻找加工路径 最优解,并按照BL 定位策略完成对矩形的排样。通过对遗传算法的改进：①对初始个体基因 位的合法性判断,并利用深度优先遍历结果评估个体性能的优劣;②交叉、变异算子均采用自 适应机制,并且执行变异操作的对象限定为一条染色体上的断点集,极大提高了算法的性能。 最后,通过实验验证了该算法在绝大多数情况下完全可以找到满足需求目标的结果,是一种非 常可靠的方法。     

Curve fitting and optimal interpolation for CNC machining under confined error using quadratic B-splines

In CNC machining, fitting the polyline machining tool path with parametric curves can be used for smooth tool path generation and data compression. In this paper, an optimization problem is solved to find a quadratic B-spline curve whose Hausdorff distance to the given polyline tool path is within a given precision. Furthermore, adopting time parameter for the fitting curve, we combine the usual two stages of tool path generation and optimal velocity planning to derive a one-step solution for the CNC optimal interpolation problem of polyline tool paths. Compared with the traditional decoupled model of curve fitting and velocity planning, experimental results show that our method generates a smoother path with minimal machining time.     

Rough and finish tool-path generation for NC machining of freeform surfaces based on a multiresolution method

Progressive fitting and multiresolution tool path generating techniques are proposed in this paper, by which multi-level (LOD) models fitting for different subsets of sampled points are obtained, and then multiresolution rough-cut and finish-cut tool paths are generated based on the LOD models. The advantages of the proposed method are: (1) the user need not care for data reduction in CAD modeling; (2) final result is obtained by interpolating two lower-level reconstructed surfces, and each lower multiresolution CAD representation can be used to generate rough-cut tool paths; (3) different manufacturing requirements utilize different level models to generate tool paths; (4) selective refinement can be applied by interpolating selceted areas at different levels of details. The key avantage of the prograssive fitting algorithm is that it can use different level surfaces to generate adaptive rough-cut and finishi-cut tool path curves directly. Therefore, based on the proposed techniques, tool path length is reduced. Sharp concers are smothed out and large tools can be selected for rough machining. The efficiency of this algorithm has been demonstrated, and it results in a 20% reduction in machining time.