There was a problem of great distraction in traditional touch interactive operation of in-vehicle information interface, but the application of gesture recognition technology on in-vehicle system improved this problem. However, users had less experience in this new interactive mode, and the cognitive deviation of the agreement between gestures and commands could directly affect the safety of drivers. The main purpose of this paper was to obtain the user's preference mid-air gestures for the in-vehicle information interface by using the user-elicitation method. In addition, the optimized gesture recognition network structure was applied to the prototype system of in-vehicle information interaction with user-defined mid-air gestures developed by us, and the effectiveness of the system was evaluated by collecting eye movement indicators of users through the simulation of driving eye movement experiments. In the process of user elicitation, the principle of command prompt and agreement rate was introduced. According to the agreement rate elicited by users (AR = .397), we got the gesture consensus set design direction of in-vehicle information control. The experimental results of the eye movement index showed that the method of user-defined mid-air gestures can effectively improve driving safety on in-vehicle media control, and reduced the distraction of users when driving compared with the traditional touch-based method. Article highlights The user's preferred in-vehicle media control task gestures were obtained through the user elicitation method. Constructed an in-vehicle secondary task control prototype system based on gesture recognition. The eye movement experiment proved that the prototype system can reduce the user's distraction when operating the in-vehicle's secondary tasks.

针对传统外骨骼研发中普遍忽略外骨骼对人体作用的缺陷,提出一种基于OpenSim开源生物力学仿真平台的外骨骼设计方法,以提高外骨骼的安全性和有效性.将外骨骼机械本体嵌入人体肌肉骨骼模型中进行耦合仿真,分析在外骨骼助力下人体特定部位关节肌肉的状态;在仿真平台中修改外骨骼的结构参数,观察其对人体助力的影响,不断迭代改进,最终得到外骨骼产品;以踝关节保护器为例,阐述外骨骼设计方法的流程.仿真分析结果表明:使用该方法设计的踝关节保护器能使人在30°斜坡上竖直着陆时脚踝距下关节内翻角度减小34％.

In this paper, we present a novel learnable and continuous Monte-Carlo Tree Search method, named as KB-Tree, for motion planning in autonomous driving. The proposed method utilizes an asymptotical PUCB based on Kernel Regression (KR-AUCB) as a novel UCB variant, to improve the exploitation and exploration performance. In addition, we further optimize the sampling in continuous space by adapting Bayesian Optimization (BO) in the selection process of MCTS. Moreover, we use a customized Graph Neural Network (GNN) as our feature extractor to improve the learning performance. To the best of our knowledge, we are the first to apply the continuous MCTS method in autonomous driving. To validate our method, we conduct extensive experiments under several weakly and strongly interactive scenarios. The results show that our proposed method performs well in all tasks, and outperforms the learning-based continuous MCTS method and the state-of-the-art Reinforcement Learning (RL) baseline.

This paper presents a novel approach to simulate assisted human walking to evaluate the assistance of the powered lower-limb exoskeleton. We first construct a subject specific musculoskeletal model as an agent that generates muscle forces according to internal and external states, and describe the exoskeleton as a rigid multi-link structure with compensatory torques applied at the joints. Then we train the agent to produce walking motion using a deep reinforcement learning algorithm given recorded experimental data on a biomechanical simulator. Next, we combine the pre-trained musculoskeletal agent with the exoskeleton and perform the assisted walking simulation which takes into account the human exoskeleton dynamics. Simulated energy expenditures under different experimental conditions are compared to evaluate the assistance effectiveness of the exoskeleton. Results show that the proposed method has great potential in providing insights into the human-exoskeleton interaction.

针对产品定位过程中只考虑用户需求权重而忽略用户需求层次变化所导致的产品定位偏差问题,提出一种基于优化Kano分析的产品定位设计决策方法.通过在Kano模型中定义用户需求矢量,利用粗糙层次分析法(R-AHP)计算用户需求权重及类型对用户需求层次进行分类,以用户满意度增量为决策依据确定下一代产品的设计方向,得到最佳的产品定位.通过采血笔的设计决策为例,验证了该方法的可行性.

As surface electromyogram (sEMG) signals have the ability to detect human movement intention, they are commonly used to be control inputs. However, gait sub-phase classification typically requires monotonous manual labeling process, and commercial sEMG acquisition devices are quite bulky and expensive, thus current sEMG-based gait sub-phase recognition systems are complex and have poor portability. This study presents a low-cost but effective end-to-end sEMG-based gait sub-phase recognition system, which contains a wireless multi-channel signal acquisition device simultaneously collecting sEMG of thigh muscles and plantar pressure signals, and a novel neural network-based sEMG signal classifier combining long-short term memory (LSTM) with multilayer perceptron (MLP). We evaluated the system with subjects walking under five conditions: flat terrain at 5 km/h, flat terrain at 3 km/h, 20 kg backpack at 5 km/h, 20 kg shoulder bag at 5 km/h and 15° slope at 5 km/h. Experimental results show that the proposed method achieved average classification accuracies of 94.10%, 87.25%, 90.71%, 94.02%, and 87.87%, respectively, which were significantly higher than existing recognition methods. Additionally, the proposed system had a good real-time performance with low average inference time in the range of 3.25 ~ 3.31 ms.

In this paper, we propose a deep neural network based approach for the group-level emotion recognition in 6th Emotion Recognition in the Wild Challenge (EmotiW 2018). The task of this challenge is to classify a group's perceived emotion as Positive, Neutral or Negative. Like the most of current researchers on visual emotion recognition, we mainly focus on facial, scene and body clues in images. We treat each clue as mono-model feature and apply early fusion method to combine them together. Experimental results show that our proposed method has outperformed the baseline techniques with the overall test accuracy of 62.90%.

With the increasing richness of people's cultural life, the density of learning gymnastics, dance, fitness and other sports activities gradually increase, and more and more users will pay more attention to the accomplishment of dance art and exercise of gymnastics. However, owing to the improper practice, many students have muscle strain, ligament tear, dislocation of muscles and bones and other injuries, which may need a long-term recovery at later stage, and some injuries may even last a lifetime to influence the physical and mental health of students. Therefore, the author makes user cards, finds and concludes user pain points, excavates function requirements, and uses KJ analysis methods, KANO model to establish product function model through researches on user background based on the study of TRIZ theory. The user problems are solved and the function requirements are realized through the exploration and application of TRIZ theory, matter-field model analysis and other methods, so as to explore the practice method conductive to students, design the exoskeleton robot as dance realia, and provide scientific and effective auxiliary training and safety protection for users.It is crucial to improve the quality of dance, gymnastics and other sports education.

Music can trigger human emotion. This is a psychophysiological process. Therefore, using psychophysiological characteristics could be a way to understand individual music emotional experience. In this study, we explore a new method of personal music emotion recognition based on human physiological characteristics. First, we build up a database of features based on emotions related to music and a database based on physiological signals derived from music listening including EDA, PPG, SKT, RSP, and PD variation information. Then linear regression, ridge regression, support vector machines with three different kernels, decision trees, k-nearest neighbors, multi-layer perceptron, and Nu support vector regression (NuSVR) are used to recognize music emotions via a data synthesis of music features and human physiological features. NuSVR outperforms the other methods. The correlation coefficient values are 0.7347 for arousal and 0.7902 for valence, while the mean squared errors are 0.023 23 for arousal and 0.014 85 for valence. Finally, we compare the different data sets and find that the data set with all the features (music features and all physiological features) has the best performance in modeling. The correlation coefficient values are 0.6499 for arousal and 0.7735 for valence, while the mean squared errors are 0.029 32 for arousal and 0.015 76 for valence. We provide an effective way to recognize personal music emotional experience, and the study can be applied to personalized music recommendation.

The recognition of human emotions from facial expression images is one of the most important topics in the machine vision and image processing fields. However, recognition becomes difficult when dealing with non-frontal faces. To alleviate the influence of poses, we propose an encoder-decoder generative adversarial network that can learn pose-invariant and expression-discriminative representations. Specifically, we assume that a facial image can be divided into an expressive component, an identity component, a head pose component and a remaining component. The encoder encodes each component into a feature representation space and the decoder recovers the original image from these encoded features. A classification loss on the components and an $\ell _{1}$ pixel-wise loss are applied to guarantee the rebuilt image quality and produce more constrained visual representations. Quantitative and qualitative evaluations on two multi-pose datasets demonstrate that the proposed algorithm performs favorably compared to state-of-the-art methods.

Adaptation to user's impedance properties has been viewed imperative in developing assistive exoskeletons. However, most exoskeletons are still rather controlled with predefined joint trajectories. Inspired by the success in locomotion control for humanoid robots using central pattern generator (CPG) and impedance control for human-robot interaction, an impedance control based on CPG that can adapt to human's impedance properties is presented in this work. The proposed method is validated by a simulation in MATLAB/Simulink, integrating a 6-DOF lower limb exoskeleton model with a musculoskeletal human model. Simulation results indicate that the lower limb exoskeleton, incorporating the proposed adaptive CPG-based impedance controller, can effectively assist human walking.