JOURNAL OF BEIJING INSTITUTE OF TECHNOLOGY

Human-Robot Interface for Unmanned Aerial Vehicle via a Leap Motion

Mingxuan Chen, Caibing Liu, Guanglong Du, Ping Zhang

2019, 28(1): 1-7. doi:10.15918/j.jbit1004-0579.18013

[Abstract](684) [PDF 3994KB](477)

Abstract:
The unmanned aircraft vehicles industry is in the ascendant while traditional interaction ways for an unmanned aerial vehicle (UAV) are not intuitive enough. It is difficult for a beginner to control a UAV, therefore natural interaction methods are preferred. This paper presents a novel interactive control method for a UAV through operator's gesture, and explores the natural interaction method for the UAV. The proposed system uses the leap motion controller as an input device acquiring the gesture position and orientation data. It is found that the proposed human-robot interface can track the movement of the operator with satisfactory accuracy. The biggest advantage of the proposed method is its capability to control the UAV by just one hand instead of a joystick. A series of experiments verified the feasibility of the proposed human-robot interface. The results demonstrate that non-professional operators can easily operate a remote UAV by just using this system.

Context-Aware Animation Data Description and Interaction Method Based on Sketch

Fang Liu, Zhaoxu Sun, Cuixia Ma, Hongan Wang

2019, 28(1): 8-16. doi:10.15918/j.jbit1004-0579.18020

[Abstract](581) [PDF 6053KB](332)

Abstract:
In this paper, an interactive method is proposed to describe computer animation data and accelerate the process of animation generation. First, a semantic model and a resource description framework (RDF) are utilized to analyze and describe the relationships between animation data. Second, a novel context model which is able to keep the context-awareness was proposed to facilitate data organization and storage. In our context model, all the main animation elements in a scene are operated as a whole. Then sketch is utilized as the main interactive method to describe the relationships between animation data, edit the context model and make some other user operations. Finally, a context-aware computer animation data description system based on sketch is generated and it also works well in animation generation process.

Lung Nodule Image Retrieval Based on Convolutional Neural Networks and Hashing

Yan Qiang, Xiaolan Yang, Juanjuan Zhao, Qiang Cui, Xiaoping Du

2019, 28(1): 17-26. doi:10.15918/j.jbit1004-0579.18022

[Abstract](525) [PDF 2759KB](390)

Abstract:
Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer. In recent years, binary hashing has become a hot topic in this field due to its compressed storage and fast query speed. Traditional hashing methods often rely on high-dimensional features based hand-crafted methods, which might not be optimally compatible with lung nodule images. Also, different hashing bits contribute to the image retrieval differently, and therefore treating the hashing bits equally affects the retrieval accuracy. Hence, an image retrieval method of lung nodule images is proposed with the basis on convolutional neural networks and hashing. First, a pre-trained and fine-tuned convolutional neural network is employed to learn multi-level semantic features of the lung nodules. Principal components analysis is utilized to remove redundant information and preserve informative semantic features of the lung nodules. Second, the proposed method relies on nine sign labels of lung nodules for the training set, and the semantic feature is combined to construct hashing functions. Finally, returned lung nodule images can be easily ranked with the query-adaptive search method based on weighted Hamming distance. Extensive experiments and evaluations on the dataset demonstrate that the proposed method can significantly improve the expression ability of lung nodule images, which further validates the effectiveness of the proposed method.

Interactive System for Video Summarization Based on Multimodal Fusion

Zheng Li, Xiaobing Du, Cuixia Ma, Yanfeng Li, Hongan Wang

2019, 28(1): 27-34. doi:10.15918/j.jbit1004-0579.18023

[Abstract](583) [PDF 3978KB](651)

Abstract:
Biography videos based on life performances of prominent figures in history aim to describe great men's life. In this paper, a novel interactive video summarization for biography video based on multimodal fusion is proposed, which is a novel approach of visualizing the specific features for biography video and interacting with video content by taking advantage of the ability of multimodality. In general, a story of movie progresses by dialogues of characters and the subtitles are produced with the basis on the dialogues which contains all the information related to the movie. In this paper, JGibbsLDA is applied to extract key words from subtitles because the biography video consists of different aspects to depict the character's whole life. In terms of fusing keywords and key-frames, affinity propagation is adopted to calculate the similarity between each key-frame cluster and key-words. Through the method mentioned above, a video summarization is presented based on multimodal fusion which describes video content more completely. In order to reduce the time spent on searching the interest video content and get the relationship between main characters, a kind of map is adopted to visualize video content and interact with video summarization. An experiment is conducted to evaluate video summarization and the results demonstrate that this system can formally facilitate the exploration of video content while improving interaction and finding events of interest efficiently.

Embedded BCI Rehabilitation System for Stroke

Wanzeng Kong, Siman Fu, Bin Deng, Hong Zeng, Jianhai Zhang, Shijie Guo

2019, 28(1): 35-41. doi:10.15918/j.jbit1004-0579.18048

[Abstract](480) [PDF 1262KB](312)

Abstract:
In stroke rehabilitation, rehabilitation equipments can help with the training. But traditional equipments are not convenient to carry, which limits patients to use related rehabilitation techniques. To solve this kind of problem, a new embedded rehabilitation system based on brain computer interface(BCI) is proposed in this paper. The system is based on motor imagery (MI) therapy, in which electroencephalogram(EEG) is evoked by grasping motor imageries of left and right hands, then collected by a wearable device. The EEG is transmitted to a Raspberry Pie processing unit through Bluetooth and decoded as the instructions to control the equipment extension. Users experience the limb movement through the visual feedback so as to achieve active rehabilitation. A pilot study shows that the user can control the movement of the rehabilitation equipment through his mind, and the equipment is convenient to carry. The study provides a new way to stroke rehabilitation.

Evaluation of Product Placement with Attention on Eye-Tracking and EEG

Wanzeng Kong, Xinyu Zhang, Luyun Wang, Qiaonan Fan, Yuanming Dai, Yunxi Miao

2019, 28(1): 42-50. doi:10.15918/j.jbit1004-0579.18028

[Abstract](692) [PDF 1330KB](388)

Abstract:
Advertising evaluation can not only evaluate the subconscious behavior of consumers and help consumers to avoid irrational consumption, but also improve the utilization of advertising resources. To deal with this topic, electroencephalogram (EEG) data and eye movement data from 18 subjects is collected. The power spectra of beta band and theta band are used to calculate the EEG attention index, while the eye attention index is calculated by the statistics at fixation on the interesting area. Therefore the fusion attention index is obtained by fusing the EEG signal and eye movement data in the feature level. The results indicate that the memory of the advertisements is synced up with the attention of the products. And the fusion attention index has a higher significant difference than the other two indexes.

Edge-Preserving Depth Map Super-Resolution with Intensity Guidance

Xiaochuan Wang, Xiaohui Liang

2019, 28(1): 51-56. doi:10.15918/j.jbit1004-0579.18019

[Abstract](429) [PDF 7164KB](263)

Abstract:
Depth discontinuity edge affects the visual quality of synthesized images in 3D image warping. However, it suffers from accuracy degradation when up-sampled from low-resolution depth maps, especially at large scaling factors. To preserve the accuracy of depth discontinuity, a novel joint bilateral depth super-resolution with intensity guidance method is proposed. Particularly, the fast local intensity classification is exploited to estimate depth coefficients in joint bilateral up-sampling for depth maps, so as to eliminate depth discontinuity edge misalignment. Additionally, the proposed method is accelerated on graphic processing units (GPUs) to meet the requirement of real-time application. Experiments demonstrate that our method can preserve the accuracy of depth discontinuity edges after super resolution, leveraging the visual quality of synthesized image in 3D image warping.

Fast and Stable Surface Feature Simulation for Particle-Based Fluids

Xiaokun Wang, Yanrui Xu, Xiaojuan Ban, Pengfei Ye

2019, 28(1): 57-66. doi:10.15918/j.jbit1004-0579.18026

[Abstract](480) [PDF 4057KB](289)

Abstract:
In order to efficiently and realistically capture microscopic features of fluid surface, a fast and stable surface feature simulation approach for particle-based fluids is presented in this paper. This method employs a steady tension and adhesion model to construct surface features with the consideration of the adsorption effect of fluid to solid. Molecular cohesion and surface area minimization are appended for surface tension, and adhesion is added to better show the microscopic characteristics of fluid surface. Besides, the model is integrated to an implicit incompressible smoothed particle hydrodynamics (SPH) method to improve the efficiency and stability of simulation. The experimental results demonstrate that the method can better simulates surface features in a variety of scenarios stably and efficiently.

EZCalm: A User Emotion Regulation System by Immersive Music Based on Plutchik's Wheel of Emotions

Yajie Xu, Lu Wang, Yulin Wang, Jiyue Hu, Peng Shao, Yijia Ma, Xiangxu Meng

2019, 28(1): 67-74. doi:10.15918/j.jbit1004-0579.18024

[Abstract](638) [PDF 2388KB](315)

Abstract:
Nowadays, many companies hire professional trainers to keep their employees emotionally positive. In addition, more and more people are visiting psychological doctors to adjust their emotion, in order to treat mental illness or gain sense of happiness. All kinds of people mentioned above want to adjust emotions in an effective way. Therefore, a simple and effective household pattern of emotion adjustment is particularly important for people to use in common days. For such a pattern, an emotion regulation pattern combined with Plutchik's wheel of emotions is proposed for emotion adjustment. In order to enhance the effectiveness of emotion adjustment from both the vision channel and the hearing channel, we adjust the user's emotion by visualizing music on a Helmet mounted display (HMD) device. Findings demonstrate that the proposed system can calm down participants' emotions effectively.

Working Longer and Happier: Inclusive Design for the Ageing Workforce

Ting Zhang, Guoying Lu

2019, 28(1): 75-82. doi:10.15918/j.jbit1004-0579.18018

[Abstract](457) [PDF 2360KB](469)

Abstract:
This paper estimates the population to be influenced by China's policy of postponed retirement age and provides design strategies and suggestions in the workplace for the ageing workforce from the perspective of inclusive design. First, the literature review about western design principles and practices for the ageing workforce is conducted. It is estimated that China would face approximately 100 million elderly workers in the near future with diverse physical conditions and demands. However, the design research and practices dedicated to ageing workforce in the workplace are comparatively weak in China. Inclusive design, one of the design approaches mainly focusing on ageing problems in developed countries, as well as its theory and methodology, could act as a good reference for Chinese policymakers and designers. Then based on the concept of inclusive design, a human-centered design model is established and design suggestions from four aspects are presented. Finally, two cases are further discussed to illustrate the idea of inclusive design.

Semi-Direct Visual Odometry and Mapping System with RGB-D Camera

Xinliang Zhong, Xiao Luo, Jiaheng Zhao, Yutong Huang

2019, 28(1): 83-93. doi:10.15918/j.jbit1004-0579.17149

[Abstract](464) [PDF 4851KB](379)

Abstract:
In this paper a semi-direct visual odometry and mapping system is proposed with a RGB-D camera, which combines the merits of both feature based and direct based methods. The presented system directly estimates the camera motion of two consecutive RGB-D frames by minimizing the photometric error. To permit outliers and noise, a robust sensor model built upon thet-distribution and an error function mixing depth and photometric errors are used to enhance the accuracy and robustness. Local graph optimization based on key frames is used to reduce the accumulative error and refine the local map. The loop closure detection method, which combines the appearance similarity method and spatial location constraints method, increases the speed of detection. Experimental results demonstrate that the proposed approach achieves higher accuracy on the motion estimation and environment reconstruction compared to the other state-of-the-art methods. Moreover, the proposed approach works in real-time on a laptop without a GPU, which makes it attractive for robots equipped with limited computational resources.

Sensitivity Analysis, Determination and Optimization of Granite RHT Parameters

Hongchao Li, Yong Chen, Dianshu Liu, Lei Zhao, Greg You

2019, 28(1): 94-102. doi:10.15918/j.jbit1004-0579.18083

[Abstract](446) [PDF 2679KB](268)

Abstract:
The RHT model has 34 parameters, among which 19 parameters can be obtained by experiments or theoretical calculations and the remaining 15 parameters are difficult to acquire. In this study, firstly, 10 Hopkinson impact tests were conducted to acquire the typical stress-strain curves of granite under dynamic loads. Through the sensitivity analysis, it is found that 13 of the 15 difficult-acquired parameters are effective to affect the shape of the stress-strain curve, and the other two parameters have no effect. Following the initial determination of model parameters with reference to the concrete RHT model, a new approach is proposed to optimize the 13 influential parameters through the LS-DYNA numerical simulation and orthogonal experiments. Finally, the determined granite RHT model parameters are verified by the results of Hopkinson impact tests conducted in this study and the bullet penetration test by Wang et al. Both results of the numerical simulations are in a good agreement with the tested results, which validates the suitability of the proposed method to acquire RHT model parameters for granite and the other rocks.

Ultra-Low Power Small Size 5.8.GHz RF Transceiver Design for WiMAX/4G Applications

Muhammad Yasir Faheem, Shun'an Zhong, Abid Ali Minhas, Muhammad Basit Azeem

2019, 28(1): 103-108. doi:10.15918/j.jbit1004-0579.180102

[Abstract](672) [PDF 2261KB](309)

Abstract:
A state of the art ultra-low power small sized transceiver design has been proposed. This device consists of four blocks, including a frequency synthesizer (FS), a crystal oscillator (XO), transmitter and receiver attached with an antenna. It has been seen that wireless information technology and systems have played a vital role in the transformation of society in different aspects of life. Mobile wireless communications including WiMAX/4G have attracted researchers and developers. WiMAX/4G applications need a transceiver that can be used in the worst channel conditions, but with low power consumption and low input voltage at the 5.8 GHz frequency. The proposed transceiver operates on 1.2 V. The operating frequency, noise figure (NF) and receiver gain are 5.8 GHz, 4.0 dB and 90 dB respectively. It is a highly compatible transceiver with all the 4 G technologies. Implementation details and results have revealed that the proposed transceiver is much more efficient than the previously proposed transceivers in literature.

Homomorphic Hashing Verification for Wireless Sensor Networks Rateless Codes Over-the-Air Programming

Hao He, Weidong Yi, Ming Li, Yongrui Chen

2019, 28(1): 109-118. doi:10.15918/j.jbit1004-0579.17150

[Abstract](406) [PDF 541KB](246)

Abstract:
The homomorphic hash algorithm (HHA) is introduced to help on-the-fly verify the vireless sensor network (WSN) over-the-air programming (OAP) data based on rateless codes. The receiver calculates the hash value of a group of data by homomorphic hash function, and then it compares the hash value with the receiving message digest. Because the feedback channel is deliberately removed during the distribution process, the rateless codes are often vulnerable when they face security issues such as packets contamination or attack. This method prevents contaminating or attack on rateless codes and reduces the potential risks of decoding failure. Compared with the SHA1 and MD5, HHA, which has a much shorter message digest, will deliver more data. The simulation results show that to transmit and verify the same amount of OAP data, HHA method sends 17.9% to 23.1% fewer packets than MD5 and SHA1 under different packet loss rates.

Interpolation Fitting Algorithm in Time-Space Domain of Differential Barometric Altimetry

Zhengqun Hu, Lirong Zhang, Guanyi Ma, Yuanfa Ji

2019, 28(1): 119-128. doi:10.15918/j.jbit1004-0579.17155

[Abstract](441) [PDF 1429KB](260)

Abstract:
In order to overcome the limitations of a unitary reference station in mobile communication positioning network differential barometric altimetry (DBA) and broaden the action scope of the reference station and improve positioning accuracy of elevation, an integrated interpolation algorithm model based on generalized extended approximation (GEA) algorithm and Kriging interpolation in time-space domain of reference station is proposed. In the time domain, barometric measured data is considered the maximum value estimated by bilateral extension to avoid wrong direction of estimation, which is approaching true value. In the spatial domain, barometric relevance among multiple reference stations is utilized, the weighted coefficients of multiple reference stations is calculated by the integrated algorithm model based on the GEA algorithm and Kriging interpolation. The impact of each reference station to the measured station is quantified, so that a virtual reference station is constructed, which can overcome the limitation of barometric correction by a unitary reference station. In addition, the measurement error due to irregular change in atmospheric pressure will be eliminated.

Feature-Based Fusion of Dual Band Infrared Image Using Multiple Pulse Coupled Neural Network

Yuqing He, Shuaiying Wei, Tao Yang, Weiqi Jin, Mingqi Liu, Xiangyang Zhai

2019, 28(1): 129-136. doi:10.15918/j.jbit1004-0579.17165

[Abstract](442) [PDF 2573KB](393)

Abstract:
To improve the quality of the infrared image and enhance the information of the object, a dual band infrared image fusion method based on feature extraction and a novel multiple pulse coupled neural network (multi-PCNN)is proposed. In this multi-PCNN fusion scheme, the auxiliary PCNN which captures the characteristics of feature image extracting from the infrared image is used to modulate the main PCNN, whose input could be original infrared image. Meanwhile, to make the PCNN fusion effect consistent with the human vision system, Laplacian energy is adopted to obtain the value of adaptive linking strength in PCNN. After that, the original dual band infrared images are reconstructed by using a weight fusion rule with the fire mapping images generated by the main PCNNs to obtain the fused image. Compared to wavelet transforms, Laplacian pyramids and traditional multi-PCNNs, fusion images based on our method have more information, rich details and clear edges.

Wideband MIMO Radar Waveform Optimization Based on Range Profile

Qun Zhang, Yishuai Gong, Yijun Chen

2019, 28(1): 137-145. doi:10.15918/j.jbit1004-0579.17132

[Abstract](448) [PDF 4445KB](299)

Abstract:
Aiming at the signal bandwidth design problem for multi-target imaging task, a kind of multiple input multiple output (MIMO) radar waveform design method is proposed. At first, the closed-loop feedback between the range profile and the signal bandwidth, which can design the minimum bandwidth of a transmitting signal that can distinguish each scatterer of the target in range direction, is established. Then, considering the request of beam pattern and the bandwidth limitation, a waveform optimization model is established and solved. Therefore, the multi-target observation and the dynamic adjustment of the signal bandwidth are accomplished. In the end, the simulation results prove the performance of the algorithm in a low SNR circumstance.

Numerical Simulation Study on Wear of Spur Gears under Dynamic Conditions

Changsong Zheng, Zhouli Zhang, Chunping He, Weipeng Lou, Huizhu Li, Qiu Du

2019, 28(1): 146-154. doi:10.15918/j.jbit1004-0579.17127

[Abstract](454) [PDF 497KB](233)

Abstract:
A numerical simulation model is proposed to predict the wear depth of gears, where Archard's wear equation and a nonlinear dynamic model are combined to establish a wear calculation model under dynamic conditions. The dynamic meshing force, determined by the non-linear dynamic model, and the sliding coefficient are used by Archard's wear equation to calculate the surface wear. Then the dynamic meshing force and sliding coefficient would be recalculated according to the surface wear state. After repeated iterations, the simulation results show that the peak and fluctuation of the meshing force increase first, then decrease, and eventually maintain stability during the process of wear. As for the distribution of wear depth, its fluctuation also increases first and then declines. Finally, the distribution of wear depth becomes V-shaped. Comparing the trends of the two factors, it is clear that the meshing force and wear depth are closely related. Moreover, the wear rate maintains a higher constant value first and then declines to a lower constant value.

Cloud Computing Based Optimal Driving for a Parallel Hybrid Electric Vehicle

Jie Fan, Yuan Zou, Zehui Kong, Ludger Heide

2019, 28(1): 155-161. doi:10.15918/j.jbit1004-0579.17160

[Abstract](431) [PDF 1075KB](305)

Abstract:
A cloud computing based optimal driving method is proposed and its feasibility is validated through a real-world scenario simulation. Based on principles of vehicle dynamics, the driving optimization problem has been formulated into an optimal control problem constrained by traffic rules, directed at achieving lower equivalent fuel consumption and shorter travel time. In order to conveniently specify the constraints and facilitate the application of the dynamic programming (DP) algorithm, the driving optimization problem is transformed into spatial domain and discretized properly. Considering the heavy computational costs of the DP algorithm, a cloud computing based platform structure is proposed to solve the optimal driving problem in real-time. A case study is simulated based on a real-world traffic scenario in Matlab. Simulation results demonstrate that the cloud computing framework is promising toward realizing the real-time energy management for hybrid electric vehicles.

Variable Parameter Self-Adaptive Control Strategy Based on Driving Condition Identification for Plug-In Hybrid Electric Bus

Kongjian Qin, Yu Liu, Xi Hu

2019, 28(1): 162-170. doi:10.15918/j.jbit1004-0579.17156

[Abstract](445) [PDF 2420KB](305)

Abstract:
A variable parameter self-adaptive control strategy based on driving condition identification is proposed to take full advantage of the fuel saving potential of the plug-in hybrid electric bus (PHEB). Firstly, the principal component analysis (PCA) and the fuzzy c-means clustering(FCM)algorithm is used to construct the comprehensive driving cycle, congestion driving cycle, urban driving cycle and suburban driving cycle of Chinese urban buses.Secondly, an improved particle swarm optimization (IPSO) algorithm is proposed, and is used to optimize the control parameters of PHEB under different driving cycles, respectively. Then, the variable parameter self-adaptive control strategy based on driving condition identification is given.Finally, for an actual running vehicle, the driving condition is identified by relevance vector machine (RVM), and the corresponding control parameters are selected to control the vehicle. The simulation results show that the fuel consumption of using the variable parameter self-adaptive control strategy is reduced by 4.2% compared with that of the fixed parameter control strategy, and the feasibility of the variable parameter self-adaptive control strategy is verified.

Analysis and Optimization of Passenger Side Knee Slider Compression in Front Impact

Chengyue Jiang, Ke Wang, Hongyun Li, Lihai Ren, Shuang Lu, Shanjun Ke

2019, 28(1): 171-175. doi:10.15918/j.jbit1004-0579.17141

[Abstract](446) [PDF 1169KB](338)

Abstract:
The influences of different design factors, as well as dummy posture, on an occupant's knee slider compression, were studied in this paper. Based on the vehicle geometry data, the simulation model, including both the multi-rigid-body and finite element (FE) part, was built up and validated with China New Car Assessment Program (C-NCAP) full impact to ensure the accuracy of the model. By adjusting the design parameters and the posture of the femur and lower leg, different factors affecting the passenger's knee slider compression were evaluated, with the help of MAthematical DYnamic MOdel (MADYMO) simulations. The study indicated that the leg posture, the stiffness of the IP and angles of the carpet have significant effects on the knee slider compression in this case. By decreasing the angle between the femur and lower leg from 133°to 124°, the maximum knee slider compression was decreased by 17.3% and by scaling the IP stiffness from 1 to 0.7, it could be decreased by 18.6%. Also, decreasing the angles of the carpet from 28°to 37° can help reduce the knee slider compression by 18.3%.

Out-of-Plane Compressive Behavior for UHMWPE/ Polyurethane Composites after Hygrothermal Treatment

Libao Zhu, Yongqing Li, Xi Zhu, Zixu Zhu

2019, 28(1): 176-183. doi:10.15918/j.jbit1004-0579.17157

[Abstract](390) [PDF 2289KB](260)

Abstract:
Quasi-static and high strain rate compressive behaviors and failure mechanisms of hygrothermal treated ultra-high molecular weight polyethylene/polyurethane (UHMWPE/PU) composites have been studied in this paper. Firstly, the UHMWPE composites were immersed in water at 70 ℃. The out-of-plane compression test was then performed on the dry/wet state specimens at quasi-static states (0.001-0.01 s^-1) and high strain rate states (800-2 400s^-1). The split Hopkinson pressure bar (SHPB) was adopted in the dynamic tests and waveform shapers were used to smooth and control the incident pulse. The results show that there are two platforms for the water absorption curve of UHMWPE composites. The absorption of moisture reduces the quasi-static compressive strength of the material while initially increasing, then decreasing the dynamic compressive strength. Matrix plasticization, fiber/matrix interface degradation and void expansion are the main factors affecting the irregular change of static/dynamic compressive strength of UHMWPE composites.

Compressive Behaviour and Failure Mechanisms of GFRP Composite at High Strain Rates

Dejun Yin, Jian Zheng, Chao Xiong, Junhui Yin, Huiyong Deng, Xiaobo Su

2019, 28(1): 184-190. doi:10.15918/j.jbit1004-0579.18075

[Abstract](609) [PDF 4688KB](342)

Abstract:
Experimental investigations into the compressive behavior of glass fiber reinforced polymer (GFRP) composite at high strain rates were carried out using a split Hopkinson pressure bar (SHPB) setup. The GFRP laminates were made from E-glass fibers and epoxy resins by vacuum assisted compression molding machine. The results of the compressive tests indicated that the mechanical behavior of the GFRP composite depends highly on the strain rate. The compressive peak stress, toughness and Young's modulus of the GFRP composite increased with the increase of strain rate, while the strain level at the initial stages of damage was shortened with the increase of strain rate. In addition, the dynamic deformation behavior and failure process of the specimens were observed directly by using a high-speed camera. Following the experiments, the fracture morphologies and damage modes were examined by scanning electron microscopy (SEM) to explore the possible failure mechanisms of the specimens. The results showed that multiple failure mechanisms appeared, such as matrix crack, fiber-matrix debonding, fiber failure and shear fracture.

2019 Vol. 28, No. 1