基于FPGA的仿人假手控制系统设计

为了满足生机电一体化仿人假手的控制需求,提出了基于FPGA的仿人假手电气控制系统设计方案。采用模块化设计思想,由FPGA构成的主控芯片模块便于功能拓展与二次开发;由DSP构成的手指运动控制模块、肌电信号采集模块、电刺激模块、USG接口模块和电池管理模块均可独立工作,与主控芯片模块间通过通用接口连接。系统集成度高,可完全放置于假手内部。应用该控制系统在HIT V代手上进行多指抓取实验,实验结果证明其工作效果良好。     

手功能康复机器人的鲁棒控制研究

以手功能康复机器人为研究对象,对被控对象建立数学模型,推导鲁棒控制器,基于simMechanics建立仿真模型,实现手功能康复机器人的鲁棒控制系统。考虑到手功能康复机器人是帮助偏瘫患者进行康复训练并最大限度地诱发患肢的主动作用,控制器中加入偏瘫患者的主动作用函数。控制系统在MATLAB／Simulink的环境下仿真,结果表明所设计的鲁棒控制器能适应不同程度的偏瘫患者,具有较佳的动态特性和很强的鲁棒性能。     

基于手势识别的机器人人机交互技术研究

研究了基于视觉的动态手势识别技术,采用基于肤色的高斯模型与改进的光流场跟踪算法结合的方 法,实现了复杂背景下实时的手势跟踪,具有快速和准确的特点,且具有较好的鲁棒性．对于动态手势识别器,采 用了隐马尔可夫模型（HMM）作为训练识别算法．考虑到动态手势特征本身的一些特点,对HMM 参数优化算法重 估式加以修正,调整了算法比例因子,从而推导了最佳状态链的确定算法、HMM 参数优化算法．最后将研究开发 的动态手势识别算法成功地应用到了基于网络的远程机器人控制系统中．     

Effects of the handle diameter and tip angle of chopsticks on the food-serving performance of male subjects

The effects of chopsticks handle diameter and tip angle on the food-serving performance of pinching food, pulling food, shearing food and thrusting food, were investigated in this study. A total of 24 male subjects was tested using 12 pairs of experimental chopsticks, consisting of three types of different handle diameters and four types of different tip angles. These results indicated that chopstick handle diameter and tip angle have a significant influence on eating efficiency, and that these two variables have a significant interaction. In addition, chopstick tip diameter also had significant effects on performance at the four tasks and subjective ratings. Generally, according to the results, when the chopsticks design is presented in terms of handle diameter, tip angle and tip diameter, a pair of chopsticks with 6 mm handle diameter × 2 ° tip angle × 4 mm tip diameter would be optimum.     

A heuristic approach to automatic grasp planning for a 3-fingered hand

This paper presents the core of a software system able to determine a good grasp configuration on 3D objects for a three-fingered hand. The grasp planning problem has been studied considering both the constraints due to the stability and accessibility conditions, and the ones related to functionality. Physical, geometrical, spatial and task-related knowledge for solving the grasp planning problem have been properly modelled to support a heuristic-based reasoning process. A series of heuristic rules and geometric tests are used to scan the solution space, searching for a good grasp. In fact, when considering the three-dimensional case, a purely analytical and exhaustive approach appears too complex because of the dimension of the search space. This approach results in an incremental and modular model of grasp reasoning, that has been implemented using the Flex expert system shell. This work has been developed and demonstrated within the Esprit 2 project CIM-PLATO No. 2202.     

A two stream convolutional neural network with bi-directional GRU model to classify dynamic hand gesture

Dynamic hand gesture recognition is still an interesting topic for the computer vision community. A set of feature vectors can represent any hand gesture. A Recurrent Neural Network (RNN) can recognize these feature vectors as a hand gesture that analyzes the temporal and contextual information of the gesture sequence. Thus, we proposed a hybrid deep learning framework to recognize dynamic hand gestures. In the Hybrid model GoogleNet is pipelined with a Bidirectional GRU unit to recognize the dynamic hand gesture. Dynamic hand gestures consist of many frames, and features of each frame need to be extracted to get the temporal and dynamic information of the performed gesture. As RNN takes input as a sequence of feature vectors, we extract features from videos using pretrained GoogleNet. As Gated Recurrent Unit is one of the variants of RNN to classify the sequential data, we created a feature vector that corresponds to each video and passed it to the bidirectional GRU (BGRU) network to classify the gestures. We evaluate our model on four publicly available hand gesture datasets. The proposed method performs well and is comparable with the existing methods. For instance, we achieved 98.6% accuracy on Northwestern University Hand Gesture(NWUHG), 99.6% on SKIG, 99.4% on Cambridge Hand Gesture (CHG) datasets respectively. We performed our experiments on DHG14/28 dataset and achieved an accuracy of 97.8% with 14-gesture classes and 92.1% on 28-gesture classes. DHG14/28 dataset contains skeleton and depth data, and our proposed model used depth data and achieved comparable accuracy.     

Performance Analysis of Machine Learning Algorithms for Classifying Hand Motion-Based EEG Brain Signals

Brain-computer interfaces (BCIs) records brain activity using electroencephalogram (EEG) headsets in the form of EEG signals; these signals can be recorded, processed and classified into different hand movements, which can be used to control other IoT devices. Classification of hand movements will be one step closer to applying these algorithms in real-life situations using EEG headsets. This paper uses different feature extraction techniques and sophisticated machine learning algorithms to classify hand movements from EEG brain signals to control prosthetic hands for amputated persons. To achieve good classification accuracy, denoising and feature extraction of EEG signals is a significant step. We saw a considerable increase in all the machine learning models when the moving average filter was applied to the raw EEG data. Feature extraction techniques like a fast fourier transform (FFT) and continuous wave transform (CWT) were used in this study; three types of features were extracted, i.e., FFT Features, CWT Coefficients and CWT scalogram images. We trained and compared different machine learning (ML) models like logistic regression, random forest, k-nearest neighbors (KNN), light gradient boosting machine (GBM) and XG boost on FFT and CWT features and deep learning (DL) models like VGG-16, DenseNet201 and ResNet50 trained on CWT scalogram images. XG Boost with FFT features gave the maximum accuracy of 88%.     

欠驱动手动态抓取控制的研究

基于手指动力学模型建立了速度观测器.该观测器用于欠驱动手的自适应轨迹跟踪,弥补了欠驱动手没有速度传感器的缺点,补偿了手指模型中欠驱动元件(如扭簧)造成的不确定因素.与PID和计算力矩法相比,其跟踪误差更小,动态控制效果更加理想.通过动力学模型建立了基于力的阻抗控制策略,在阻抗控制中使用速度观测器,实现了基关节的动态抓取力跟踪.动态轨迹跟踪与力跟踪相结合,使欠驱动手在抓握时具有良好的动态抓取性能.     

融合深度与肤色特征的自适应手部跟踪算法

手部跟踪技术是实现自然人机交互的关键。针对现有跟踪方法易受光照、环境等影响及鲁棒性差的不足,提出一种融合深度与肤色特征的自适应手部跟踪算法。考虑手部运动过程的形变,该算法首先利用深度平滑连续性选取深度阈值以实现跟踪区域的自适应尺度变化,获得手部候选区域。在此基础上建立YCbCr空间肤色归一化直方图,在粒子滤波框架下将跟踪问题转换为贝叶斯估计问题,基于最大后验准则确定手部位置,并通过监测粒子重要性权值的方差解决跟踪失效问题,实现复杂观测环境下的鲁棒跟踪。实验结果表明,该跟踪算法可适应不同复杂环境,鲁棒性良好。