基于双向搜索的改进蚁群路径规划算法

针对移动机器人在复杂地图环境中移动耗时长、易陷入局部最优等问题,设计了一种基于双向搜索的改进蚁群路径规划算法。基于K-means算法对地图预处理,量化地图的局部复杂度程度,并将局部环境信息融合到状态转移概率函数,使机器人优先选择在复杂程度小的区域进行寻优,减少路径拐点。设定双向搜索规则,改进启发函数,提高算法的局部方向搜索精度和全局搜索效率。针对蚁群算法中蚂蚁遇到U障碍物陷入死锁的问题,提出死锁判断系数,增加了有效蚂蚁的数量,进一步提高了算法性能。仿真结果表明所设计的算法在复杂地图环境中相较于传统蚁群算法移动机器人的路径搜索效率更高。     

面向三维结构视觉检测的无人机覆盖路径规划

为了高效地规划无人机在三维覆盖检测任务的飞行路径,建立了满足覆盖率要求的路径规划模型,可分为两步：第1步确定无人机巡检的视点和视线,第2步计算视点的无碰撞访问序列。首先,从巡检目标的三维点云出发,提出基于K均值聚类生成候选视点的方法,构建候选视点互连的非完全图模型;其次,提出面向排序的混合蚁群算法(sorting-oriented hybrid ant colony algorithm, S-HACO)求取无人机巡检路径,优化目标考虑了路径长度、视点数目、急转弯次数等。仿真实验结果表明,该方法得到的视点数目较偏移法和随机采样法分别减少了96.25%和42.10%,并且S-HACO算法较传统算法相比性能更优,目标函数降低了19.14%,算法的运行时间减少了25.27%,验证了模型的有效性和算法的可行性。     

Neural reactive path planning with Riemannian motion policies for robotic silicone sealing

Due to its excellent chemical and mechanical properties, silicone sealing has been widely used in many industries. Currently, the majority of these sealing tasks are performed by human workers. Hence, they are susceptible to labor shortage problems. The use of vision-guided robotic systems is a feasible alternative to automate these types of repetitive and tedious manipulation tasks. In this paper, we present the development of a new method to automate silicone sealing with robotic manipulators. To this end, we propose a novel neural path planning framework that leverages fractional-order differentiation for robust seam detection with vision and a Riemannian motion policy for effectively learning the manipulation of a sealing gun. Optimal control commands can be computed analytically by designing a deep neural network that predicts the acceleration and associated Riemannian metric of the sealing gun from feedback signals. The performance of our new methodology is experimentally validated with a robotic platform conducting multiple silicone sealing tasks in unstructured situations. The reported results demonstrate that compared with directly predicting the control commands, our neural path planner achieves a more generalizable performance on unseen workpieces and is more robust to human/environment disturbances.     

一种面向低采样率的点云数据处理网络

在直接处理点云的三维神经网络中,采样阶段实现了对原始点云中关键点的筛选,对于整个网络的性能及网络的抗噪能力具有重要作用。目前主流的最远点采样（FPS）方法在处理大规模3D点云数据时计算量大且耗时,并且低采样率时经过FPS采样后模型性能下降明显。针对这两个问题,提出一种面向低采样率的点云数据处理网络AS-Net。设计一个新的采样模块代替原backbone中的FPS,其由两个Layer组成,每个Layer基于长短期记忆网络获取原始点云与采样点云之间的联系权重,从而高效提取关键信息,去除冗余信息。在此基础上,利用注意力机制选择特征值较高的原始点云作为采样点,采样点作为后序任务的关键点输入到网络,进一步提高网络模型性能。基于ModelNet40数据集的实验结果表明,在低采样率条件下,AS-Net仍可达到81.6%的分类准确率,与使用FPS作为采样方法的网络模型相比提高52.7%。此外,其对噪声干扰具有很强的鲁棒性,对于大场景的分割时间效率优于同类采样方法。     

Rule-based meta-heuristics for integrated scheduling of unrelated parallel machines,batches, and heterogeneous delivery trucks

In this article, we deal with an integrated scheduling for unrelated parallel machines, batches, and heterogeneous delivery trucks. In a manufacturing plant, jobs ordered by customers are manufactured by one of several unrelated parallel machines. Then, they are grouped and delivered to the respective customers by heterogeneous trucks with different capacities and travel times. The objective of the problem is to simultaneously determine the machine schedule, batching, and truck-delivery schedule to minimize the make span of the entire process. To solve this problem, we derive a mathematical model to obtain the optimal solution, and we propose rule-based meta-heuristics using single-stage GA framework. Through randomly generated instance examples, the performances of the proposed meta-heuristics are compared.     

A novel database of children's spontaneous facial expressions (LIRIS-CSE)

Computing environment is moving towards human-centered designs instead of computer centered designs and human's tend to communicate wealth of information through affective states or expressions. Traditional Human Computer Interaction (HCI) based systems ignores bulk of information communicated through those affective states and just caters for user's intentional input. Generally, for evaluating and benchmarking different facial expression analysis algorithms, standardized databases are needed to enable a meaningful comparison. In the absence of comparative tests on such standardized databases it is difficult to find relative strengths and weaknesses of different facial expression recognition algorithms. In this article we present a novel video database for Children's Spontaneous facial Expressions (LIRIS-CSE). Proposed video database contains six basic spontaneous facial expressions shown by 12 ethnically diverse children between the ages of 6 and 12 years with mean age of 7.3 years. To the best of our knowledge, this database is first of its kind as it records and shows spontaneous facial expressions of children. Previously there were few database of children expressions and all of them show posed or exaggerated expressions which are different from spontaneous or natural expressions. Thus, this database will be a milestone for human behavior researchers. This database will be a excellent resource for vision community for benchmarking and comparing results. In this article, we have also proposed framework for automatic expression recognition based on Convolutional Neural Network (CNN) architecture with transfer learning approach. Proposed architecture achieved average classification accuracy of 75% on our proposed database i.e. LIRIS-CSE.     

基于具有墙角信息的语义地图改进AMCL重定位算法

针对原始自适应蒙特卡洛定位（Adaptive monte carlo localization,AMCL）算法仅利用激光信息存在的缺陷,提出一种基于激光与视觉融合的语义地图进行全局定位,该语义地图融合基于深度学习的目标检测方法提取环境中的墙角语义;利用建立的包含墙角信息的二维语义栅格地图,结合视觉预定位方法及角点周围语义信息表来提高算法全局初始定位的效率和准确性,使得移动机器人可以在少量先验信息和运动的情况下更迅速地实现定位。提出视觉预定位的方法,改进了粒子权重更新方式,再同步结合AMCL算法与环境地图匹配进行精定位。最后通过搭建的移动机器人在不同场景下进行对比试验,验证了该方法的有效性。     

结合历史运动状态的机器人高效沿墙算法研究

针对目前室内移动机器人沿墙走算法过于复杂、路径易重复、不能完全遍历、效率低等问题, 采用室内未知环境下结合历史状态的机器人沿墙高效遍历研究来解决这些问题. 该算法由移动机器人的上一个周期历史环境运动状态(分8类)、当前环境运动状态(分8类)和旋向信息(分2类)建立运动规则库, 沿墙行走时移动机器人时时采集这三类信息(上一个周期历史环境运动状态、当前环境运动状态和旋向信息)决定移动机器人当前的运动方向, 如此循环直到完成指定的沿墙任务. 最后对该算法进行了仿真与实际实验, 实验结果证明该算法可以在不同的、复杂的环境中高效、快速地完成沿墙走的任务, 并且对室内未知环境有很好的适应性.     

基于变结构ESKF的航姿参考系统噪声处理方法

针对航姿参考系统（AHRS）易受到环境与传感器自身噪声干扰,导致姿态估计精度下降的问题,提出了一种基于变结构误差状态卡尔曼滤波（VS-ESKF）的噪声数据处理方法。首先,通过分析AHRS传感器观测数据与新息序列统计特征,设计了基于加速度范数与遗忘序贯概率比检验（F-SPRT）的方法,分别检测加速度计与陀螺仪的噪声数据。其次,基于噪声检测结果,将平滑变结构滤波(SVSF）策略引入到误差状态卡尔曼滤波（ESKF）,以提高ESKF对噪声模型不确定性的处理能力。然后,结合磁场强度与磁倾角参数特征,利用马氏距离法评估磁干扰并实时调整磁力计补偿权重,获取准确的AHRS修正数据。最后,基于自主搭建独轮机器人平台进行实验验证,结果表明所设计的VS-ESKF算法可以及时、准确地检测AHRS噪声数据,并有效地抑制噪声干扰,相比于ESKF算法,对横滚角、俯仰角和偏航角的估计精度分别提升了31.05%、32.32%和40.07%,提高了姿态估计的准确性和稳定性。     

Design,fabrication, and testing of fiber Bragg grating sensors for cryogenic long-range displacement measurement

It is essential to measure the shrinkage/expansion and positioning/aligning of magnets and to control valve displacement which plays a vital role in experiments like the Karlsruhe tritium neutrino experiment beam tube and Cryo pumps. Hence, a displacement sensor which, over a long working range, can be operated under extreme environmental conditions needs to be developed. Fiber Bragg gratings (FBG) have been considered to be excellent sensor elements useful for a variety of applications. This paper will discuss a long range displacement sensors based on fiber Bragg gratings for cryogenic temperature applications. The cryo pump inlet valve control requirements have been taken as example specifications for sensor design. To achieve the development goal, a proper signal transducer and sensor package were designed. A study of the strain transmission of surface-bonded FBG was conducted. The influence of bonding thickness and bonding length was reported. The design, fabrication, and performance were tested at low temperature of around 77 K. The sensor performance was found to be satisfactory at both room temperature and 77 K and linearly for long-range displacement of 550 mm with 14 pm/mm sensitivity and 0.142 mm accuracy.