Motion Planning by T‐RRT with Potential Function for Vertical Articulated Robots

RRT (rapidly exploring random tree) with random sampling is an effective method for path planning, and is often used for robot manipulators. The RRT has many modified methods for applying various problems and conditions. Particularly, T‐RRT (Transition‐based RRT) one of those has advantage that it is able to adopt arbitrary evaluation function. In this paper, a novel path planning method based on the T‐RRT is proposed for ensuring “quality” of a generated path. Then, its effectiveness is evaluated via comparison with other sampling‐based methods using simulation of the industrial robot having seven DOFs.     

A rapid microwave-assisted derivatization process for the determination of phenolic acids in brewer’s spent grains

A simple and fast method for the TMS derivatization of phenolic acids in a closed vial using microwave irradiation followed by GC/MS analysis is proposed. A full factorial design was used to investigate the effects of two independent variables, namely, time and power of microwave irradiation, on the yield of silylation reaction. The optimal conditions were tested against the classical heating derivatization procedure. No significant differences were found between the classical heating and microwave-assisted derivatization. Chromatographic separation of the nine phenolic acids examined was achieved in 16 min. The mass spectral fragmentation patterns of the derivatives obtained by the proposed method were identical to those from the classical heating. Four different batches of brewer’s spent grain were extracted and analyzed for the total phenolic acid content. Significant differences between the batches of spent grains were found for all analytes. The total phenolic acid content varied between 2688 and 4884 μg/g.     

基于改进RRT路径规划算法

针对基本RRT算法路径规划特点:树的扩展具有随机性,路径中存在冗余的节点,规划出的路径拐角多。因此,提出一种改进的RRT算法。改进后的算法首先采用目标偏向策略,以一定的概率P把目标点作为采样点进行随机树扩展,提高随机树向目标点的扩展的概率;其次,改变度量函数,添加了角度约束,减少了路径规划中搜索的范围,以减少搜索时间;最后,通过贪心算法对路径进行简化,并对简化后的节点采用三次B样条曲线进行优化。结果证明改进后的RRT算法搜索时间更少,搜索路径更短,平滑性更高。     

面向改进RRT算法的回转空间布线技术研究

为解决回转空间下曲率变化大、线缆搜索效率低等问题,提出一种面向改进RRT算法的线缆路径规划的方法。通过对回转空间建模和布线空间划分进行研究,减少了不必要的搜索空间,提高了搜索效率。为解决RRT算法中采样点和扩展方向随机性过强的问题,避免节点"斜跨"表面敷设过大,提出了轴向约束角度采样策略,同时为提高算法整体搜索效率,提出动态步长扩展策略。最终通过实例测试与其他传统算法进行对比,证明了该算法的有效性和可行性。     

无先验地图条件下电力管廊无人机自主巡检方法研究

针对无先验地图条件下电力管廊无人机自主巡检能力不足的问题,本文提出了一种不依赖先验地图的无人机巡检方法。首先,无人机搭载激光雷达,通过SLAM（Simultaneous Localization and Mapping）算法进行实时场景构图,并基于RRT（Rapidly-exploring Random Tree）算法对构建的全局栅格地图进行边界搜索;其次,针对传统RRT算法难以判断环境地图完整性、保障巡检全面性的问题,提出了一种基于数字图像处理的边界检测方法。通过Canny算子对实时构建的地图进行边缘检测,并对地图进行完整性评估。最后,为验证本文提出方法的有效性,在江苏无锡处某220kV电力管廊的仿真模型中进行验证,分别采用RRT算法和本文改进算法进行自主巡检测试,结果表明本文提出的算法相比RRT算法可提升21.8%的巡检覆盖率。     

动态环境下基于人工势场引导的RRT路径规划算法

现有的大多数动态RRT路径规划算法不能使规划的路径远离障碍物,这有可能导致机器人没有足够的避障时间。针对此问题,提出了一种利用人工势场引导快速扩展随机树向目标区域生长并远离障碍物的改进RRT算法APFG-RRT(artificial potential field guided RRT)。为了进一步加快算法的收敛速度、加速算法跳出局部极小值,引入了一种按自适应概率选择目标点作为采样点的策略;针对动态环境采用全局规划结合局部重新规划的方法以提高算法的实时性。仿真实验表明,相比于初始RRT和Goal-bias RRT,APFG-RRT的计算效率更高,内存需求更小,并且搜索到的路径能够有效地远离障碍物,提高了动态路径规划的成功率。     

低空复杂环境下基于采样空间约减的无人机在线航迹规划算法

针对低空复杂环境下障碍物密集且类型多样、带有多通道并存在不确定信息的无人机在线航迹规划问题,为了减少碰撞检测次数,提高航迹搜索速度,降低航迹代价,提出一种基于采样空间约减的无人机在线航迹规划算法. 算法通过引入代价模型,提出约减域逐步构造方法,引导规划树快速有效扩展,改善了基于动态域的快速拓展随机树（Dynamic domain rapidly-exploring random tree,DDRRT） 算法中存在的采样空间过度约减问题. 算法通过密度划分索引的方法逐步构建多棵Kd 树（K-dimensional tree）并采用多近邻节点搜索方法,加快了近邻树节点搜索速度. 仿真实验结果表明,与DDRRT方法相比,该方法在保证对采样空间约减合理性的同时,提高了航迹规划效率和通道内的寻路能力.     

Fluid overload and acute kidney injury

Acute kidney injury is commonly encountered in critically ill patients, and is associated with worse outcomes. Fluid therapy is a key component in the management of these patients, often leading to fluid overload, especially in the setting of septic acute kidney injury. Emerging data overwhelmingly suggest that fluid overload in these patients may be associated with adverse outcomes. Management of such patients should include a strategy of early guided resuscitation, followed by careful assessment of fluid status, and early initiation of renal replacement therapy as soon as it is deemed safe, aiming for a neutral or negative fluid balance. This review will focus on the pathophysiological link between fluid overload and acute kidney injury, mechanisms of organ dysfunction in fluid overload, and strategies for management.     

分级随机采样弱随机RRT算法及在移动机器人运动规划中的应用

针对移动机器人运动规划过程中,采用快速扩展随机树(RRT)算法采样效率低,寻找临近节点计算量大,及非线性反馈控制器不受系统模型动态约束的问题, 提出一种新的基于分级随机采样与扩展的弱随机RRT算法,同时设计快速限幅非线性反馈控制器,保证运动规划过程中机器人始终满足系统模型动态约束。首先,在迭代伊始结合节点评价策略建立节点的选取集合;其次,按照规定顺序选取扩展节点并随机选择扩展方向,将计算得到的新子节点连接到随机树完成扩展;然后,对初始路径进行规划,采用快速限幅非线性反馈控制器计算机器人在路径点上的控制序列和位姿,实现移动机器人的运动规划;最后,通过仿真验证了该算法的有效性。结果表明:提出的分级随机采样弱随机RRT算法不依赖最近节点的选取,相比RRT算法缩短了求解时间,提高了迭代速度。