NOVEL APPROACH FOR ROBOT PATH PLANNING BASED ON NUMERICAL ARTIFICIAL POTENTIAL FIELD AND GENETIC ALGORITHM

A novel approach for collision-free path planning of a multiple degree-of-freedom （DOF） articulated robot in a complex environment is proposed. Firstly, based on visual neighbor point （VNP）, a numerical artificial potential field is constructed in Cartesian space, which provides the heuristic information, effective distance to the goal and the motion direction for the motion of the robot joints. Secondly, a genetic algorithm, combined with the heuristic rules, is used in joint space to determine a series of contiguous configurations piecewise from initial configuration until the goal configuration is attained. A simulation shows that the method can not only handle issues on path planning of the articulated robots in environment with complex obstacles, but also improve the efficiency and quality of path planning.

Path planning for unmanned aerial vehicles in surveillance tasks under wind fields

The optimal path planning for fixed-wing unmanned aerial vehicles(UAVs) in multi-target surveillance tasks(MTST) in the presence of wind is concerned.To take into account the minimal turning radius of UAVs,the Dubins model is used to approximate the dynamics of UAVs.Based on the assumption,the path planning problem of UAVs in MTST can be formulated as a Dubins traveling salesman problem(DTSP).By considering its prohibitively high computational cost,the Dubins paths under terminal heading relaxation are introduced,which leads to significant reduction of the optimization scale and difficulty of the whole problem.Meanwhile,in view of the impact of wind on UAVs' paths,the notion of virtual target is proposed.The application of the idea successfully converts the Dubins path planning problem from an initial configuration to a target in wind into a problem of finding the minimal root of a transcendental equation.Then,the Dubins tour is derived by using differential evolution(DE) algorithm which employs random-key encoding technique to optimize the visiting sequence of waypoints.Finally,the effectiveness and efficiency of the proposed algorithm are demonstrated through computational experiments.Numerical results exhibit that the proposed algorithm can produce high quality solutions to the problem.

Multi-Behavior Fusion Based Potential Field Method for Path Planning of Unmanned Surface Vessel

The problem of the unmanned surface vessel (USV) path planning in static and dynamic obstacle environments is addressed in this paper. Multi-behavior fusion based potential field method is proposed, which contains three behaviors: goal-seeking, boundary-memory following and dynamic-obstacle avoidance. Then, different activation conditions are designed to determine the current behavior. Meanwhile, information on the positions, velocities and the equation of motion for obstacles are detected and calculated by sensor data. Besides, memory information is introduced into the boundary following behavior to enhance cognition capability for the obstacles, and avoid local minima problem caused by the potential field method. Finally, the results of theoretical analysis and simulation show that the collision-free path can be generated for USV within different obstacle environments, and further validated the performance and effectiveness of the presented strategy.

Machine learning inspired workflow to revise field development plan under uncertainty

We present an efficient and risk-informed closed-loop field development (CLFD) workflow for recurrently revising the field development plan (FDP) using the accrued information. To make the process practical, we integrated multiple concepts of machine learning, an intelligent selection process to discard the worst FDP options and a growing set of representative reservoir models. These concepts were combined and used with a cluster-based learning and evolution optimizer to efficiently explore the search space of decision variables. Unlike previous studies, we also added the execution time of the CLFD workflow and worked with more realistic timelines to confirm the utility of a CLFD workflow. To appreciate the importance of data assimilation and new well-logs in a CLFD workflow, we carried out researches at rigorous conditions without a reduction in uncertainty attributes. The proposed CLFD workflow was implemented on a benchmark analogous to a giant field with extensively time-consuming simulation models. The results underscore that an ensemble with as few as 100 scenarios was sufficient to gauge the geological uncertainty, despite working with a giant field with highly heterogeneous characteristics. It is demonstrated that the CLFD workflow can improve the efficiency by over 85% compared to the previously validated workflow. Finally, we present some acute insights and problems related to data assimilation for the practical application of a CLFD workflow.

Ant Colony Optimization with Potential Field Based on Grid Map for Mobile Robot Path Planning

For the mobile robot path planning under the complex environment,ant colony optimization with artificial potential field based on grid map is proposed to avoid traditional ant colony algorithm's poor convergence and local optimum.Firstly,the pheromone updating mechanism of ant colony is designed by a hybrid strategy of global map updating and local grids updating.Then,some angles between the vectors of artificial potential field and the orientations of current grid are introduced to calculate the visibility of eight-neighbor cells of cellular automata,which are adopted as ant colony's inspiring factor to calculate the transition probability based on the pseudo-random transition rule cellular automata.Finally,mobile robot dynamic path planning and the simulation experiments are completed by this algorithm,and the experimental results show that the method is feasible and effective.

动态目标的Field D＊算法及路径的提取计算

FieldD*算法计算的路径消耗较优并较为平滑,但该方法的计算量很大,尤其是在对动态目标的规划时,计算时间无法满足实际要求。针对该缺陷,在对FieldD*计算原理分析的基础上,提出了在目标是动态的情况下,只进行前目标与新目标的路径规划,并对FieldD*计算所得路径消耗只进行局部修改的动态规划算法,该动态算法可以有效减少计算量。将得到的路径消耗值应用到路径的提取算法中,路径提取算法针对路径消耗的不同情况下,提出不同的路径提取方法,并最终得出总体路径消耗最优的路径。在以栅格法建立的数字海图中进行仿真验证,该动态计算和路径提取算法与D*算法相比,计算所得的路径更为平滑,路径的总体消耗也更少,缺点是所需计算时间略长。

基于运动约束的泛化Field D~＊路径规划

为了解决基于栅格的路径规划算法因环境描述的离散化导致规划结果不能满足机器人运动约束,以及单一路径代价的局限致使算法无法适用于复杂环境的问题,提出一种基于运动约束的泛化Field D*算法.该算法的代价函数可同时考虑路程、行驶安全以及行驶时间等一个或多个行驶代价.根据机器人运动模型的特性,在路径点提取过程中结合机器人的最小转弯半径,进行满足运动约束的路径平滑.该算法在多组模拟的复杂环境栅格地图中进行测试,实验结果表明,算法对复杂环境有很好的适应性,同时有效提高路径的可执行性.

改进A^(*)算法和人工势场法的路径规划

A^(*)算法存在折线路径多和搜索节点多的问题,人工势场(artificial potential field,APF)法存在局部最优和不可到达的问题,针对两种算法存在的问题进行了研究。利用欧氏距离与投影距离提出一种新的混合式启发函数,依据该函数对A^(*)算法的流程进行改进,减少A^(*)算法的搜索节点,提高搜索效率。利用新A^(*)算法生成的最优节点作为APF算法的局部目标点,辅助机器人摆脱局部最优点;通过加入机器人和目标点的位置关系改进势场函数,修改斥力的增益,优化斥力的生成方向。在改进的基础上将两种算法融合提出一种新的算法,利用APF法的势场函数引导A^(*)算法的搜索。从路径长度、避障效果、迭代次数对改进算法进行对比分析,仿真结果表明,提出的改进算法搜索效率高,实现避障的同时保证计算的路径最优。

Design and analysis of an advanced thermal management system for the solar close observations and proximity experiments spacecraft

In this paper,the mission and the thermal environment of the Solar Close Observations and Proximity Experiments(SCOPE)spacecraft are analyzed,and an advanced thermal management system(ATMS)is designed for it.The relationship and functions of the integrated database,the intelligent thermal control system and the efficient liquid cooling system in the ATMS are elaborated upon.For the complex thermal field regulation system and extreme space thermal environment,a modular simulation and thermal field planning method are proposed,and the feasibility of the planning algorithm is verified by numerical simulation.A solar array liquid cooling system is developed,and the system simulation results indicate that the temperatures of the solar arrays meet the requirements as the spacecraft flies by perihelion and aphelion.The advanced thermal management study supports the development of the SCOPE program and provides a reference for the thermal management in other deep-space exploration programs.

改进人工势场算法的路径规划

人工势场法是一种适用于局部路径规划的算法,针对人工势场法存在的目标不可达、局部极小值等固有缺陷,提出了一种改进后的人工势场法。首先针对目标不可达问题,将引力影响因子添加到斥力场的生成中,改进斥力场函数,从而避免引力与斥力合力为零的情况。针对局部极小值问题,通过设立虚拟目标点来引入额外外力,打破机器人的平衡问题。通过与其他算法的对比实验,仿真结果显示,改进人工势场法规划的路径长度和消耗时间都更短,稳定性更强。

胸中上段食管癌患者调强放疗照射野数的剂量学研究

目的:比较5野、7野IMRT用于胸中上段食管癌患者调强放射治疗(IMRT)的剂量学差异。方法:纳入接受IMRT的150例胸中上段食管癌患者为研究对象,根据IMRT照射野数不同分为7野组(n=71)及5野组(n=79)。全部患者总治疗剂量为50 Gy,每周期治疗5次,共计治疗25次。比较两组治疗期间靶区及相关危及器官剂量学参数,比较两组治疗前后血清学指标及对治疗的满意度。结果:两组患者靶区平均受照剂量比较,差异无统计学意义(P>0.05);7野组患者均匀指数低于5野组(P<0.05),适形指数高于5野组(P<0.05)。两组患者心脏、脊髓、左肺、右肺平均受照剂量比较,差异均无统计学意义(P>0.05);7野组患者左肺及右肺V5均高于5野组(P<0.05),左肺及右肺V20、V30均低于5野组(P<0.05)。治疗后,两组患者血清血管内皮生长因子(VEGF)、基质金属蛋白酶-2(MMP-2)、MMP-9水平均低于治疗前(P<0.05),且7野组均低于5野组(P<0.05)。两组患者满意度比较,组间差异无统计学意义(P>0.05)。结论:5野IMRT并未增加胸中上段食管癌患者相关危及器官的受照剂量,且有利于降低双肺的V5,但7野IMRT用于在靶区适形度及临床疗效方面更具优势。

Clinical Experience and Dosimetry Outcome in Treating Breast Cancer with Field-in-Field Technique

Purpose: The study is aimed to establish the dosimetric characteristics of field-in-field (FiF) technique for carcinoma of breast treatment in Nepal. We assumed that FIF technique may result in improved dose distribution and reduced acute toxicity in these patients. Methods: Forty breast cancer patients participated in this study. A total dose of 50 Gy in 25 fractions was prescribed to the planning target volume. FiF plan was generated in treatment planning system. Dose volume histograms were evaluated for PTV and organs at risks. Several parameters were analyzed for the PTVs and organ at risks (OARs) together with the Conformity index (CI), and the Homogeneity index (HI). Results: The dose coverage of breast volume was achieved. The V95% (volume of 95%) of PTV covered D95% (Dose of 95%). The PTV dose was covered to 49.98 ± 0.9 Gy and 49.81 ± 1.1 Gy for the left and right breast, respectively. The mean lung dose was 14.87 ± 0.91 Gy. The homogeneity index (0.26 ± 0.17 and 0.22 ± 0.13) and conformity index (1.59 ± 0.75 and 1.36 ± 0.45) were analyzed for left and right breast, respectively. Conclusion: The study supports the viability of FiF technique in the treatment of breast cancer in Nepal. The FIF technique enables better dose distribution in the PTV and reduces dose to OARs. The FiF technique provides dosimetric advantages and requires less planning time.

基于改进蚁群算法的农业机器人多田块路径规划方法与试验

[目的]针对丘陵地区因田块碎片化和障碍物影响农业机器人作业的情况,提出一种基于改进蚁群算法和A算法相结合的多田块路径规划方法。[方法]通过无人机构建农田遥感影像,利用栅格法对农田进行环境建模,再进行子区的划分与合并,在蚁群算法中引入区域启发函数,对不同区域内的移动代价作区分,通过人工增加最优路径中的信息素浓度并建立自适应信息素挥发系数,对信息素更新方法进行改进,以此求解作业子区间的最优遍历顺序,利用具有启发式搜索功能的A算法进行子区连接路径规划,最终实现多田块路径规划。[结果]MATLAB仿真试验结果表明,在子区数量为40时,采用改进蚁群算法进行规划的平均路径长度比传统的蚁群算法减少了3.19%,平均迭代收敛次数减少了79.5%;在路径遍历仿真试验中,农业机器人遍历路径覆盖率能达到100%,路径重复率为6.48%。利用农田无人机遥感影像和田间作业参数进行自然环境的仿真试验,进一步验证了该方法的有效性。[结论]本研究提出的方法能有效解决丘陵地区农业机器人多田块路径规划问题,可为丘陵地区农业机器人大面积作业提供路径参考与技术支持。

改进A^(*)算法的移动机器人全局路径规划

针对A^(*)算法在移动机器人路径规划存在搜索效率低,路径斜穿障碍物顶点,路径拐弯多等问题。提出一种改进的A^(*)算法,首先在A^(*)算法的邻域扩展中采用避免斜穿障碍物顶点的策略;再引入障碍物因素对评价函数进行指数加权,减少不必要的搜索,提高A^(*)算法的效率和灵活性,使算法偏向于选择障碍物较少的路径;最后使用三次优化折线的策略,加入障碍物安全距离,减少路径上的冗余节点和拐弯。使用MATLAB进行实验仿真,结果表明,在20 m×20 m、40 m×40 m、60 m×60 m栅格地图环境下,改进A^(*)算法较传统A^(*)算法,搜索时间分别减少70.12%、84.31%、91.44%,扩展节点分别减少53.77%、71.20%、74.30%,路径累计拐弯角度分别减少70.48%、76.31%、82.18%,改进A^(*)算法能够有效的提高移动机器人路径规划的效率,路径更为平滑和安全,且在复杂环境中优势更为明显。

A Linear Interpolation-Based Algorithm for Path Planning and Replanning on Girds

Field D* algorithm is widely used in mobile robot navigation since it can plan and replan any-angle paths through non-uniform cost grids. However, it still suffers from inefficiency and sub-optimality. In this article, a new linear interpolation-based planning and replanning algorithm, Update-Reducing Field D*, is proposed. It employs different approaches during initial planning and replanning respectively in order to reduce the number of updates of the rhs-values of vertices. Experiments have shown that Update-Reducing Field D* runs faster than Field D* and returns smoother and lower-cost paths.

基于改进人工势场法的AUV全局路径规划

目的为了解决AUV(autonomous underwater vehicle)在传统人工势场法路径规划过程中存在的碰撞、目标不可达和局部极小值等问题,方法提出一种基于改进人工势场法(improved artificial potential field,IAPF)的AUV全局路径规划算法。首先为解决引力过大导致的碰撞问题,引入作用阈限制目标点的吸引力;然后在斥力场添加距离修正因子修正斥力函数,防止目标点离障碍物过近时,由于斥力大于引力引起目标不可达现象,同时考虑到障碍物大小的不等性会影响斥力的作用范围,对障碍物的最小安全距离进行修改;最后在此基础上添加外力,利用外力和势场力的共同作用使AUV沿切线方向行驶,从而摆脱局部极小点或陷阱区域。结果仿真实验结果表明,本文算法能够规划出一条最优路径,与同类算法相比,本文算法运行时间短,成功率高。结论本文提出的基于改进人工势场法的AUV全局路径规划算法,通过引入作用阈、距离修正因子、最小安全距离修正以及外力作用,有效解决了传统人工势场法在AUV路径规划中存在的碰撞、目标不可达和局部极小值等问题。

基于人工磁场的防疫机器人路径规划

针对传统人工势场法在防疫机器人路径规划过程中容易陷入局部极小值和目标不可达的问题,提出一种基于人工磁场的路径规划算法。首先,对障碍物周围建立人工磁场,引入洛伦兹力,洛伦兹力与引力保持垂直,避免机器人陷入局部极小值点;其次,在洛伦兹力函数中引入机器人与目标点的距离影响因子,解决目标不可达问题;再次,通过建立目标点与障碍物之间的虚拟直线优化洛伦兹力的方向,避免迂回路径,减少路径规划的步数。最后,在MATLAB环境下对传统人工势场法和人工磁场法进行了仿真实验,实验结果表明,人工磁场法可以克服局部极小值和目标不可达问题,规划的路径对障碍物进行了避障,避免了徘徊振荡的情况出现,提高了路径规划的质量。

计及安全性与舒适性的智能车辆换道轨迹规划研究

针对当前智能车辆在换道过程中的安全性以及乘员舒适性问题,本文提出一种基于风险场评估轨迹的二次筛选方法。首先,在Frenet坐标系下,将车辆运动解耦为横向和纵向两个维度,基于五次多项式生成所有横向d-t曲线簇和纵向s-t曲线簇;其次,由车辆的动力学特性和三圆碰撞模型设计轨迹初筛评价函数,选取合格轨迹作为候选轨迹;最后,参考人工势场理论的思想,引入行车过程中风险场的概念,根据换道效率,换道风险值和横、纵向冲击度建立总损失函数评价候选轨迹以进行二次筛选,选取最佳轨迹并完成可视化。为检验算法的可行性,通过搭建双车道的道路环境,设计障碍车不同速度和加速度的多场景进行弯道换道的仿真验证。研究结果表明,本文所提算法能够满足换道的安全性和舒适性需求。同时,在正常换道场景下,乘员在换道过程的97.5%时间处于舒适状态,在紧急避障场景下也能达到平衡安全性与舒适性的目标。

基于改进人工势场算法的煤矿井下机器人路径规划

路径规划是煤矿机器人在煤矿井下狭小巷道空间中应用亟待解决的关键技术之一。针对传统人工势场(APF)算法在狭小巷道环境中规划出的路径可能离巷道边界过近,以及在障碍物附近易出现目标不可达和路径振荡等问题,提出了一种基于改进APF算法的煤矿机器人路径规划方法。参考《煤矿安全规程》有关规定建立了巷道两帮边界势场,将机器人行驶路径尽量规划在巷道中间,以提高机器人行驶安全性;在障碍物斥力势场中引入调节因子,以解决目标不可达问题;引入转角限制系数以平滑规划出的路径,减少振荡,提高规划效率,保证规划路径的安全性。仿真结果表明:当目标点离障碍物很近时,改进APF算法可成功规划出能够抵达目标点的路径;改进APF算法规划周期数较传统算法平均减少了14.48%,转向角度变化累计值平均减少了87.41%,曲率绝对值之和平均减少了78.09%,表明改进APF算法规划的路径更加平滑,路径长度更短,规划效率和安全性更高。