A method for robot placement optimization based on two-dimensional manifold in joint space

This article addresses a method for placement determination of robotic drilling system on two-dimensional manifold in robot joint space. It has been proved that the feasibility of positioning error compensation on two-dimensional manifold, and that the continuity of the robot parameters in the two-dimensional space is the prerequisite to perform the compensation in previous study. It appears that there are bifurcations which might break the continuity of the robot parameters on the two-dimensional manifold due to improper placement. To avoid bifurcations, a performance index and a set of optimization procedure are proposed to achieve proper placement of robotic machining system. Experiments conducted on a KUKA robot have verified the effectiveness of the proposed placement optimization method. Experiment results indicated that positioning errors were significantly improved with the proposed method, which is beneficial for robotic machining accuracy.     

D-2-DenseNet噪音鲁棒的城市音频分类模型

为了提高噪音环境下城市音频分类系统的鲁棒性,提出了一种双特征2阶密集卷积神经网络（D-2-DenseNet）噪音鲁棒的城市音频分类模型.首先介绍了噪音添加和噪音鲁棒处理,阐述了一种双特征互补偿的算法;然后结合2阶密集卷积神经网络与自适应机制提出了一种噪音鲁棒音频分类模型：双特征2阶密集卷积神经网络.模型采用双特征互补偿自适应算法,可在特征提取与模型训练中更有针对性地提取有效音频信息,降低噪音干扰,以提高噪音鲁棒性.最后,基于Dcase2016数据集开展噪音环境下城市音频分类测试.实验结果表明,模型分类准确率分别可达77.12%、75.52%,与基线模型相比,平均分类准确率分别提高了8.51%和10.38%,验证了模型良好的噪音鲁棒性.     

基于下肢动力学检测分析的站起康复训练机器人控制

提出了人体站起轨迹控制方法(TCM)和阻尼控制方法(ICM)。采用这两种控制方法可以实现对机器人系统的控制,使患者能安全、有效地进行下肢站起运动康复训练。在患者站起过程中,下肢各肢段旋转角、身体运动轨迹、地面反力(GRF)、压力中心(COP)和绳索张力等参数可由康复机器人系统中的相应传感器实时检测得出,并通过实时运算得出髋、膝、踝各关节力矩,用于反馈控制。试验结果表明,该方法可以保证患者实现安全、舒适、有效的站起训练,适用于站起康复训练机器人控制系统。     

基于虚拟人的汽车内饰造型草图设计方法研究

基于当前汽车内饰造型草图设计中存在的问题,引入虚拟人技术生成造型草图设计的辅助模板,建立了基于辅助模板的造型草图设计新方法,并将设计方法进行设计实践,用实际效果验证理论,从而为汽车内饰设计师提供了一种快捷而有效的技术支持方案。     

基于提取双选紧密特征的RGB-D显著性检测网络

针对现有算法对不同来源特征之间的交互选择关注度欠缺以及对跨模态特征提取不充分的问题,提出了一种基于提取双选紧密特征的RGB-D显著性检测网络。首先,为了筛选出能够同时增强RGB图像显著区域和深度图像显著区域的特征,引入双向选择模块(bi-directional selection module, BSM);为了解决跨模态特征提取不充分,导致算法计算冗余且精度低的问题,引入紧密提取模块(dense extraction module, DEM);最后,通过特征聚合模块(feature aggregation module, FAM)对密集特征进行级联融合,并将循环残差优化模块(recurrent residual refinement aggregation module, RAM)配合深度监督实现粗显著图的持续优化,最终得到精确的显著图。在4个广泛使用的数据集上进行的综合实验表明,本文提出的算法在4个关键指标方面优于7种现有方法。     

多智能体强化学习的机械臂运动控制决策研究

针对传统运动控算法存在环境适应性较差,效率低的问题。可以利用强化学习在环境中不断去探索试错,并通过奖励函数对神经网络参数进行调节的方法对机械臂的运动进行控制。但是在现实中无法提供机械臂试错的环境,采用Unity引擎平台来构建机械臂的数字孪生仿真环境,设置观察状态变量和设置奖励函数机制,并提出在该模型环境中对PPO(proximal policy optimization)与多智能体（agents）结合的M-PPO算法来加快训练速度,实现通过强化学习算法对机械臂进行智能运动控制,完成机械臂执行末端有效避障快速到达目标物体位置,并通过该算法与M-SAC（多智能体与Soft Actor-Critic结合）和PPO算法的实验结果进行分析,验证M-PPO算法在不同环境下机械臂运动控制决策调试上的有效性与先进性。实现孪生体自主规划决策,反向控制物理体同步运动的目的。     

Response map evaluation for RGBT tracking

Neural Computing and Applications - In recent years, RGB and thermal sensors are widely used. There is complementary information from these two types of sensors. A fundamental task which arises in...     

绿色气相缓蚀剂的复配及其缓蚀性能研究

目的 为解决单组分绿色气相缓蚀剂缓蚀性能差的问题,复配一种绿色复合气相缓蚀剂,探究其对碳钢和黄铜金属试样的缓蚀作用。方法 采用腐蚀质量损失、接触角、电化学等试验测试分析复合气相缓蚀剂对碳钢、黄铜的缓蚀效果与成膜耐久性。结果 复合气相缓蚀剂对10号钢、H62黄铜的缓蚀效率分别为84.71%、91.67%,缓蚀性能显著优于单组分气相缓蚀剂;复合气相缓蚀剂在10号钢、H62黄铜表面均形成了缓蚀膜,H62黄铜表面形成的缓蚀膜较10号钢的更具耐久性。结论 与单组分气相缓蚀剂相比,该复合气相缓蚀剂对碳钢、黄铜均具备良好的缓蚀作用,为绿色气相缓蚀剂的防锈包装应用提供支撑。     

PBSVM: Partitioning and biased support vector machine for vocal fold pathology assessment using labeled and unlabeled data sets

Most of the existing classification methods, used for voice pathology assessment, are built based on labeled pathological and normal voice signals. This paper studies the problem of building a classifier using labeled and unlabeled data. We propose a novel learning technique, called Partitioning and Biased Support Vector Machine Classification (PBSVM), which tries to utilize all the available data in two steps: (1) a new heuristically partition-based algorithm, which extracts high quality pathological and normal samples from an unlabeled set, and (2) a more principle approach based on biased formulation of support vector machine, which is fairly robust to mislabeling and unbalance data problem. Experiments with wavelet-based energy features extracted from sustained vowels show that the new recognition scheme is highly feasible and significantly outperform the baseline classical SVM classifier, especially in the situation where the labeled training data is small.