The application of autonomous emergency braking (AEB) is faced with challenges of misidentification and unreasonable decision. Testing under typical scenarios can effectively improve the applicability of AEB. Based on the analysis of relevant research literature, two typical scenarios (S11 and S12) involving participants turning that AEB test focuses on are extracted from 12 types of two-wheeler and vehicle pre-crash scenarios. Pre-scan model and mathematical model are established to conduct qualitative and quantitative analysis on the effectiveness and improvement direction of AEB under the typical scenarios. In scenario S11 and S12, the trajectory of the two-wheeler in the moving coordinate system is only related to the speed ratio between the vehicle and the two-wheeler. In scenario S11, only when the speed ratio of vehicle to two-wheeler is $ \left( {0.788\;8, + \infty } \right) $, the two-wheeler can enter the AEB triggering domain. By increasing the field of view from $ {60^\circ } $ to $ {90^\circ } $, the AEB can be effectively improved, and the range of speed ratio increases from $ \left( {0.788\;8, + \infty } \right) $ to $ \left( {0.203\;3, + \infty } \right) $. In scenario S12, when the speed ratio of two-wheeler to vehicle is $ (0,{{(\Delta v + 2.46)} \mathord{\left/ {\vphantom {{(\Delta v + 2.46)} {9.64}}} \right. } {9.64}}) $, the AEB has a good triggering effect. When the triggering width of AEB is expanded from 1.5 m to 2 m, the AEB triggering effect is not greatly improved. The research methods and conclusions can provide technical support for AEB optimization.

UAV swarm cooperative confrontation is a development direction for future war. In order to highlight the advantages of the swarm such as strong attack, difficult defense and high flexibility, it is an important research direction to effectively model the complex system of high-dimensional, strong dynamic and nonlinear UAV cluster cooperative confrontation. In this paper, we apply the complex spatial network theory to construct a cooperative confrontation network, a cooperative network and a confrontation network between two UAV swarms. Meanwhile, we establish a UAV swarm cooperative confrontation model in 2D and 3D based on the cooperative reconnaissance scene of UAV swarm. Then, we analyze the impact of the spatial distance between opponent UAVs on the hit rate, and put forward the formula of hit rate with spatial distance. We analyze the robustness of the cooperative network of UAV swarm through cascading effects and verify the effectiveness and practicality of the UAV swarm cooperative confrontation model. Our work will provide new insight for the modeling of UAV swarm cooperative confrontation.

In order to quantify the effect of thermal runaway gas produced in the thermal runaway process of lithium-ionbattery on thermal runaway propagation, based on the energy conservation equation and the equivalent substitution method, this paper presents a method to calculate the contribution of thermal runaway gas combustion to thermal runaway propagation. The self-designed experimental platform of thermal runaway gas energy release calculation was used to pick the commercial 18650 battery and collect the parameters required for the calculation. According to the experiment results, the energy released during the combustion of the first cell’s thermal runaway gas accounts for 5.42% of the energy needed by the second cell’s thermal runaway, resulting in a 42% increase in the second cell’s self-heat production and a 29% reduction in thermal runaway time. The research results are helpful to further explore the energy transfer efficiency in the process of runaway heat transfer, and provide theoretical support for cell level and system level battery safety design.

Lying is a complex cognitive process whose executive function requires the participation of different brain regions. And the interaction between these brain regions has been confirmed by related research. In view of the problems of limited current EEG signal feature extraction methods and the unclear psychological mechanism of lying, we constructed EEG signals-based brain networks by phase transfer entropy during the lie experiment, and we have analyzed the effective connectivity between different brain regions in the honest group and the lying group. First, the standard three-stimulus experiment was used to conduct a lying detection experiment on 60 subjects. The EEG signals of all subjects were collected simultaneously and preprocessed.Then, the phase transfer entropy was used to construct the effective connectivity matrix. Subsequently, the statistical method was used to analyze the entropy value difference between the two groups of each edge in the matrix, and the electrode pairs with significant differences in entropy value were selected as the classification features of the fully connected neural network. The result shows that the classification accuracy rate is 96.75%, indicating that it is effective to use phase transfer entropy index to distinguish the EEG signals of the liar and the honest. Finally, the brain function network of the two groups of people was analyzed. The results show that compared with honest people, there is a stronger flow of information between the frontal, parietal and temporal lobes of the liar, indicating that deception requires coordination and utilization of more brain resources. The above analysis results will help reveal the brain’s neural activity mechanism in a lying state.

As the most important core component of electric vehicles, the lithium-ion power battery system requires improvement in power performance, safety, and reliability for further large-scale development of electric vehicles in China. Thermal safety issues of lithium-ion power batteries may occur throughout the whole life cycle of the battery system, and manifest differently in different spatial scales of monomers, modules, and systems. To address the multi-spatial scale thermal safety problems of the lithium-ion power battery system, the latest progress of thermal safety design of power batteries is reviewed from three aspects: heat generation of a single battery, temperature uniformity of a module, and safety and reliability of a battery system. Some important research results are introduced, and the key problems to be solved in the thermal safety design of a lithium-ion power battery system are summarized. Feasible solutions to the problems are put forward, and the future research direction is suggested, providing insight into increase in power performance, safety and reliability of battery systems.

To address the difficulties caused by the low accuracy and poor robustness of simultaneous localization and mapping (SLAM), an ORB-SLAM2 algorithm is proposed based on key frame selection. First, the relative pose between frames is calculated based on ORB-SLAM2. Second, to determine whether a new key frame should be created, rotation and translation values are added to the original algorithm, functioning as the judgement basis. Then, an inferior key frame removal algorithm is designed to solve the problem of inferior key frame generation which results from incorrect shooting caused by the relative movement between the robot and the camera installed in the self-developed mobile robot. Finally, experiments are carried out based on the RGB-D dataset and the developed mobile robot, verifying the outstanding performance of the proposed algorithm. The results show that the improved key frame selection algorithm can accurately and timely choose the key frame, and reduce tracking failures. In the most optimal case, the positioning error is about 51.9% of that of the original, while the linear error is about 82.1% of that of the original, which effectively eliminates the influence caused by relative motion between the camera and the robot. This research shows that the improved algorithm could effectively promote positioning accuracy and reduce tracking failures.

A multi-scale real-time tracking algorithm for unmanned aerial vehicle (UAV) based on dual-attention shuffling is proposed to solve the problems of small size, large scale variation and similar object interference which often occur during UAV object tracking. First, considering the small number of target pixels in the UAV view, a deep network with double sampling integration is constructed, which provides semantic information-rich depth features and preserves the target’s detailed information. Next, a dual-attention shuffling module is designed. Channel attention and spatial attention are simultaneously grouped to filter the extracted feature information, and then the information between different channels is shuffled to enhance information exchange and improve the discriminative ability of the algorithm. Finally, to utilize the feature information of different layers, multiple region proposal networks are added to complete the target classification and regression, and the results are weighted and fused for the UAV target characteristics. Results show that the success and precision rates of the algorithm are 60.3% and 79.3% on the dataset, respectively, with 37.5 frame/s. The algorithm discrimination ability and multi-scale adaptation are significantly enhanced, which can effectively deal with the common challenges in UAV tracking.

Increasing traffic demand and the saturated terminal airspace call for safer flight performance and more efficient control performance. Therefore, this paper focuses on the multi-objective arrival sequencing and scheduling based on point merge system (PMS). Firstly, the four-dimensional trajectory prediction model was developed to predict future trajectories since trajectory prediction played a critical role in arrival sequencing and scheduling. Secondly, the PMS-based multi-objective arrival sequencing model was put forward while taking both the PMS operation modes and different stakeholders appeals into account. Moreover, the multi-objective imperialist competition algorithm (ICA) was proposed based on a non-dominated sorting strategy for solving such a problem. Finally, the data from the Monte Carlo simulation and actual operating of Changsha Huanghua International Airport were used for validation. The results indicated that the proposed method was effective and efficient. On the one hand, the proposed multi-objective ICA did well in practical application, making decision support for controllers. On the other hand, the proposed method could reduce the total delay time, total flight time, and maximum flight time by 70.8%, 13.2%, and 11.8%, compared with the actual operation, with relatively conservative safety separations.

Though the massive open online courses (MOOC) have greatly changed the way of learning, properly understanding the user’s behavior and then predication of dropout is one of the most challenging tasks. MOOC have significantly altered the way that people learn, yet one of the most difficult challenges is correctly interpreting user behavior and then predicting dropout. In this research, to improve the dropout prediction performance, we firstly analyzed users and courses from the perspective of activities by using the long short term memory mechanism. In this study, we used the long short term memory mechanism to analyze users and courses from the perspective of activities in order to improve the dropout prediction performance. Afterwards we further proposed a multi-attention based multi-perspective feature enhancement method to investigate the correlated activities among users and courses. Finally, we provided a gated mechanism-based feature integration framework for dropout prediction. The experiment study on the public dataset has shown our framework’s promising potential, thereby making it possible to better investigate the reason beneath these phenomena and improve the overall study experience. The experiment study on the open dataset has demonstrated the promising potential of our framework, allowing us to more thoroughly explore the causes of these events and enhance the learning environment as a whole.

In recent years, the application of additive manufacturing technology in manned space engineering has developed rapidly. This paper summarizes the additive manufacturing technologies used in manned space engineering, such as fused deposition modeling technology, selective laser melting technology, wire and arc additive manufacturing technology, thermal spraying additive technology, lunar soil additive manufacturing technology and their application fields. This paper also summarizes the application fields of additive manufacturing technology in on-orbit manufacturing of aircraft replacement parts, manufacturing of large truss and other components that are difficult to be manufactured or launched on the ground, and manufacturing of complex aircraft components.Finally, this study makes the case that in order to implement additive manufacturing in the future, a material system appropriate for manned space engineering must be developed. Additive manufacturing in a microgravity environment should be studied. Also, relevant processes should be developed in the future.

The inorganic-organic hybrid phenolic composite (IPC) tends to be widely used for the thermal protection of near-space hypersonic vehicles. The thermal conductivity estimation of IPCs plays an important role in the fine design of thermal protection system. A thermal conductivity identification method considering ablation effect is proposed and verified based on the benchmark of Ablation Workshop. The results show the computation precision of the proposed method. Through the arc wind tunnel test of IPCs with stratified temperature and ablation sensors, the temperature and pyrolysis thickness distribution data of IPCs with different thicknesses are obtained, and the relationship between the thermal conductivity and temperature of IPCs is achieved. Before 800 K, the thermal conductivity of the original layer of the IPC increases slowly with temperature, remaining below 0.1 W/(m·K). After 800 K, however, changes occur abruptly, and the thermal conductivity of the carbonization layer increases sharply with the increase of temperature, reaching 0.17 W/(m·K) at 1300 K.

Turboshaft engine is the main part of the power system of rotor aircraft such as helicopter. Once the key engine parameters exceed the limit, the method of reducing fuel quantity and power is generally adopted to limit, which will temporarily reduce the speed of power turbine, making it about 4%−6% lower than the normal rated state. However, if the overlimit state is not removed in time, the speed of the power turbine will continue to decrease, threatening the flight safety. Based on the flight test analysis of a certain helicopter from phenomena to data to solve the above problems. This paper proposes a control method, through the design from the total distance control law, achieve dynamic system boundary protection control in the condition of engine parameters overrun. if overrun status can’t get out timely, then engine change automatically to recover the normal control of power system. This method significantly enhances the robustness of helicopter power system control and the safety of the flight. The correctness of the design is verified by modeling the power system and simulating the control law.

In recent years, the application of unmanned impact rollers in high embankment engineering of airport has become a new trend. However, large tracking errors often occur at the start and the end point of the headland, which affects the compaction effectiveness of the working area. In this paper, a path optimization method is proposed to improve the steering tracking performance. Firstly, the U-shaped turning path derived from generalized elementary curve is established based on two calculation methods, and the optimized path is selected near the original planning path considering minimum turning radius and the smoothness of the curvature of curve. Then, the Ω-shaped turning path is formed by bi-elementary curve, and the optimized path is selected with the same method as above. Finally, the model predictive control (MPC) simulator is established to simulate the trajectory tracking effect of various paths. The results showed that the tracking effect of the optimized path is better than the original path, indicating the effectiveness of the proposed path optimization method.

In order to solve the problems of the high cost of air transportation and waste of idle transportation resources, this paper puts forward the research on route optimization of aircraft based on transshipment under route alliance. Firstly, based on the problem of cargo transfer, considering the impact of the alliance on the operation, the selection probability of aviation alliance is introduced to determine the self operation and outsourcing of segment transportation before and after the transfer, and the connection of outsourcing transportation is also considered. The aircraft route optimization model based on transshipment, known as the T-AAAFRP model, is then built in the alliance environment while taking into account the capacity limitation of double airports in the aviation network, the capacity limitation of all cargo aircraft in flight time and airspace in operation, and taking the total cost minimization as the goal Secondly, an adaptive genetic algorithm is used to solve the model. Finally, through a case study, the location and path optimization problems are studied. The results show that the algorithm designed in this paper has high convergence. In the process of changing the number of transfer points, double airport cities are always selected as transfer points. The change of demand and aircraft fixed cost has great influence on optimization decision. The decision to optimize is greatly influenced by changes in demand and aircraft fixed costs. The weight of aircraft, the sharing coefficient of alliance self operation and outsourcing, and the change of decision-maker’s risk preference have little influence on the optimization decision. But on the whole, the larger the number of transfer points, the smaller the total cost and the smaller the number of aircraft used. However, generally speaking, the more transfer sites there are, the lower the overall cost and the fewer aircraft are needed.

The security of the global navigation satellite system (GNSS) has aroused widespread concern. The multiple-antennas method has become the most effective spoofing detection method due to its unique spatial characteristics. A joint attitude determination and spoofing detection method using three antennas is proposed, detecting spoofing signals and determining the attitude information. The baseline accuracy limits the traditional direct attitude determination method, so a length-constrained baseline vector estimation method is adopted to obtain high-precision attitude determination results. When attitude information is known, the carrier phase single-difference expected value can be obtained by the ephemeris, attitude transformation matrix, and the antennas’ geometric relationship. The sum of the squared error (SSE) is used to evaluate the deviation between the observed and expected value of carrier phase single-difference, and the spoofing signal binary detection is constructed. The results shows that this method can reduce the standard deviation of attitude determination by more than 76.1% when there is no spoofing signal, and achieve 100%. detection efficiency with a more than 77.3% reducation on standard deviation of attitude when spoofing signal involved.

The model errors in the target tracking process will lead to the degraded performance and decreased filtering accuracy of the square-root cubature Kalman filter (SRCKF). Amended Kalman filter (AKF) can solve this problem effectively, but it is difficult to be applied to nonlinear filtering. To overcome the negative impact of the model errors and to further improve the application scope of the amendment thought, the vector form of the amendment coefficient is derived by minimizing innovation covariance on the basis of the SRCKF. Then, the amended SRCKF (ASRCKF) algorithm is proposed. By using posterior measurements, the ASRCKF algorithm can increase confidence level to measurement, so that the target model error can be compensated. The simulation results indicate the ASRCKF can suppress the model errors effectively with better filtering performance, compared with SRCKF and STF-SRCKF algorithms.

During the flight in space station, astronauts are prone to mental fatigue, which is the main factor that affects the efficiency of operations and causes errors. For this reason, studying rapid detection methods for human mental fatigue will help ensure the safety of on-orbit operations. The characteristic changes of the electroencephalogram (EEG) can reflect the fatigue state of the brain. Still, the existing EEG method requires multiple lead signals when analyzing mental fatigue, which seriously limits its practical application in the space station environment. This study successfully induced various mental fatigue states in 45 subjects through a foundation experiment using 36 hours of sleep deprivation. Aiming at the non-stationarity of EEG signals, the designed 8-layer db4 wavelet transform structure effectively decomposes δ, θ, α, and β brain rhythm waves. First, screen out the mental fatigue sensitivity characteristics using analysis of variance (ANOVA) and Logistic regression. Secondly, according to the number of sensitive features of mental fatigue, the sharp leads of mental fatigue were further screened out. Finally, the characteristics of 6 keen leaders were used to construct random forest regression models. Finally, the weighted fusion of the regression models at 6 leads to a rapid detection model of mental fatigue, with an average accuracy rate of up to 85.25%.

Aiming at the problems of low accuracy and under-fitting in current fault diagnosis methods for piston pumps based on deep neural networks with small samples, a new fault diagnosis method for piston pumps based on Siamese neural networks was proposed. A test bench for piston pumps was built to collect the vibration signals of the pump housing under different health states. The convolution layers and pooling layers were used to construct the Siamese sub network and adaptively extract low-dimensional features from the raw vibration signals. The similarity of the input sample pairs was determined by Euclidean distance to expand training samples, train the Siamese neural network model. And finally identify the health states on the testing dataset. Experimental results demonstrate that compared with traditional deep neural networks, the proposed method has higher diagnosis accuracy with small samples. In addition, data fusion experiments show that the proposed method can learn relevant fault information from signals in different channels, which can improve the accuracy of the fault diagnosis.

In order to solve the problem that the color image watermarking algorithm has low algorithm fault tolerance, and the size matching problem between the host image and the watermark image during embedding, and to improve the robustness of the algorithm in attack resistance, this paper proposes a robust multiple water-marking algorithm for color images based on bi-dimensional empirical mode decomposition (BEMD) and discrete cosine transform (DCT). Firstly, Arnold transform is used to scramble three binary watermark images. Then, the RGB channels of the color host image are then decomposed using BEMD in addition to derive the intrinsic modal functions (IMFs) and residues for each channel. The first IMF of each channel is selected as the watermark embedding layer and recorded as IMF1. After that, each channel is divided into non-overlapping sub-blocks, and DCT is performed on each sub-blocks. Finally, the scrambled binary watermark image is repeatedly embedded in the middle bands coefficients of each channel sub-block after a Zigzag scan, and the inverse Zigzag scan and inverse DCT are adopted to obtain the IMF1 after embedding watermark information in each channel, and then the remaining intrinsic modal functions and residues of each channel are used to reconstruct the color image after the watermark embedding. The watermark extraction is the inverse of the watermark embedding process. The algorithm in this paper can implement blind extraction of embedded watermarks. In the process of watermark extraction, the voting strategy is used to extract the repeatedly embedded watermarks, which enhances the fault tolerance of the algorithm. A large number of experimental results show that the peak signal-to-noise ratio (PSNR) of the host images is above 34 dB after several watermarks have been embedded on various sets of host photos, according to a vast number of experimental results, and the watermark images have a high degree of invisibility. The host images after embedding can be against various attacks such as large-scale cropping, salt and pepper noise, etc. The values of the extracted watermark images are all above 0.96, and some can reach 1, watermark images can be completely extracted and precisely recognized. Compared with a large number of existing color image watermarking algorithms, the color image watermarking algorithm proposed in this paper has a strong ability to resist various attacks, and the images after embedding watermarks have higher invisibility.

Carbon/glass hybrid composites have shown great potential in industrial applications. The infrared radiation characteristics of carbon/glass hybrid composite laminates and two types of non-hybrid composites under low velocity impact were studied experimentally by infrared thermography. The damage mode of the laminates was determined after impact by visual, ultrasonic C-scan and optical microscopy, and then the time series variation and temperature distribution characteristics of the thermal map sequence were analyzed to characterize the heat dissipation effect during the impact. Results show that the infrared thermography is highly suitable for monitoring the damage process of fiber reinforced composites under low velocity impact, and that the relationship between the monitoring characteristics and the damage modes can be established through the thermal map sequence. It is also found that the interlaminar hybrid of carbon glass fibers can effectively improve the anti-delamination ability of carbon fiber reinforced polymer (CFRP) composites. With the increase of impact energy, the anti-delamination ability becomes more obvious. After impact, carbon glass hybrid composites show both larger surface damage and smaller delamination damage with better damage tolerance.

To study the feasibility and potential of hydrogels for thermal management, a novel polyvinyl alcohol (PVA) /polyethylene glycol (PEG) composite hydrogel heat sink with good mechanics and economy was prepared by freezing-thaw method. The size of the sink is 60 mm×60 mm×2 mm, and self-cooling is realized by water evaporation on the surface. A heat dissipation experiment was carried out with a heat flow of 2712 W/m², and the heating characteristics, the relationship of the evaporative convection intensity change and the swelling change law were obtained. It was found that the addition of 2.5% PEG reduced the preparation deformation caused by the increase in the number of freezing cycles, with the water content attenuation being decreased by 75.53%, and the chip surface temperature 7.53%. Based on the experimental results, the evaporation heat transfer coefficient was calculated, and the effects of heat flow, thickness, and humidity on evaporative heat dissipation were studied. The swelling rates of the hydrogel with different temperatures and usage conditions (4 h continuous use and 120-day storage at room temperature) were then measured, showing that the hydrogel has a certain short-term reliability in spite of the insignificant sensitive response to temperature.

We established a Markov decision process model with two stages of long-distance autonomous guidance and short-distance companion flight avoidance for multi-rotor UAVs to track dynamic ground targets. An improved proximal policy optimization (PPO) algorithm is proposed. Considering that the data received by the UAV are time-sequential and that the environment has contextual relevance, the algorithm uses long short-term memory (LSTM) network to calculate reward values, update network parameters, and perform adaptive optimization iterations through status information such as the real-time position relationship between the UAV and the target. Experiments were conducted with a simulation testing platform based on ROS. Results show that the method proposed safely and effectively realizes autonomous maneuvering during the whole process of the reconnaissance mission. Compared with the traditional PPO algorithm, the algorithm proposed shortens the model training time due to the introduction of LSTM neural network, thus significantly improving the efficiency of obstacle tracking and avoidance. This result further strengthens the robustness, accuracy, and real-time performance of the algorithm.

Aiming at the damage problems such as debonding of lattice sandwich structure, a non-contact damage imaging technology based on high-frequency dynamic response is proposed. The high-frequency response of the structure is analysed according to the non-baseline damage index to realize debonding damage imaging. In the numerical simulation, based on the local resonance theory, the low-order natural frequency of the damage area is used as the centre frequency to calculate the response of the structure in a wide frequency range under sound field excitation, and the damage imaging is realized by using no baseline damage index. The damage imaging results can accurately identify the damage location; In the test, the non-contact test measurement scheme of using the loudspeaker to excite, and the laser vibration measurement system to pick up the full-field mode shape during scanning successfully identified the location of the debonding damage. The applicability and feasibility of the non-contact imaging technology for the damage detection of the dot matrix sandwich structure debonding is verified, and the damage identification without additional structural quality and without health reference signals is realized.

Due to the multipath effect of the radar signal, the blind source separation algorithm based on the instantaneous linear mixture model is no longer applicable. A blind source extraction method for complex non-Gaussian signals based on the FastICA algorithm is proposed. The mixed system is modeled as a convolutional linear mixture model, so that each multipath signal does not need to be regarded as an independent source signal in the signal model, which not only saves the number of receiving channels, but also reduces the complexity of blind source separation process. The non-Gaussian feature of the signal to be extracted is used to extract complex non-Gaussian sources in Gaussian background. The experimental results show that when the signal to interference ratio is −30 dB, the proposed method can quickly and effectively deal with the extraction of complex non-Gaussian sources in the convolutional linear mixture model, which provides a new method for weak signal extraction in this scene.

To deal with the problem that the square-root cubature Kalman filter (SRCKF) will have declined filtering performance or even filtering diverge when model errors and abnormal measurements occur, an adaptively robust multi-sensor fusion algorithm is proposed. Firstly, a robust subsystem is designed based on the innovation covariance matching principle to suppress abnormal measurements. Then, to overcome the model errors, an adaptive subsystem is designed based on the low complexity adaptive SRCKF (LCASRCKF) algorithm. Finally, according to the characteristics and limitations of the two subsystems, a global fusion architecture is proposed, which enables the system to fully balance and utilize the prior model prediction information and the posterior measurement information in the filtering process, and thus reduce the estimation error. The simulation results show that the proposed fusion algorithm has obvious advantages in terms of filtering accuracy, stability and convergence speed compared with the robust multiple fading factors cubature Kalman filter (RMCKF) and other algorithms.