This PDF file contains the front matter associated with SPIE Proceedings Volume 12699, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

Based on camera and millimeter wave radar data fusion, an intelligent vehicle obstacle detection method suitable for hazy environment is proposed. Firstly, by taking the effectiveness of obstacle detection after image dehazing as the evaluation standard, a series of typical dehazing networks are compared and the best one was selected for image preprocessing. An obstacle detection model based on YOLOv5s depth network was established; Then, the camera data and radar data are fused in time and space, and the sensor data is associated based on the global nearest neighbor data association algorithm. Finally, the effectiveness of the proposed method is verified by open source data sets and real vehicle experiments.

The influence of AlGaN (Al=0.1) back barrier layer and graded back barrier layer on AlGaN/GaN high electron mobility transistor (HEMT) and HEMT based sensor was investigated by Silvaco TCAD. The results show that, both the AlGaN (Al=0.1) back-barrier HEMT and the graded AlGaN back-barrier HEMT can improve drain current and transconductance compared with conventional AlGaN/GaN HEMT. The HEMT with graded back barrier layer shows larger drain current and higher transconductance than that with AlGaN back barrier, which can be attributed to that the introduction of graded AlGaN back barrier layer increases the carrier concentration at the heterojunction interface and greatly reduces the effect of the parasitic channel. Correspondingly, the device performance should exhibit stronger dependence on surface charge. To confirm the inference, the sensing performance of the HEMTs without back barrier, with Al_0.1Ga_0.9N back barrier and with graded back barrier are simulated by adding surface charges on HEMT sensing area. The results indicate that the current sensitivity ΔI_D of the graded back-barrier HEMT caused by surface charge is largest, so the HEMT with a graded back barrier layer is more suitable for ion sensor than that conventional AlGaN/GaN HEMT and fixed Al-content back barrier.

Fire accidents occur frequently in highway tunnels, and the process of fire source from generation to disaster in tunnels is very fast. It is of great significance to set fire detectors in highway tunnels to improve fire control ability, so as to ensure the safety of life and property. At present, there is still some phenomenon of failing to report or false alarm for fire detectors, and the sensing parameters and installation location of fire detectors need to be determined based on the spatial distribution pattern of fire sources to give full play to their accurate sensing function. In this paper, the spatial distribution pattern of potential fire sources in road tunnels is systematically revealed from the vehicle, road and tunnel facilities aspects, and the area within 1.5m of the tunnel headroom from the road surface is found to be the most concentrated high-risk area for potential fire source distribution. Finally, the layout of fire source sensing equipment in the high-risk areas of the tunnel and the suggestions for the emergency management of tunnel fire source sensing are put forward.

This paper proposes the linearly constrained minimum variance (LCMV) beamforming method based on steepest descent (SD) and Broyden-Fletcher-Goldfarb-Shanno (BFGS) methods for underwater acoustic beamforming. The proposed methods avoid the covariance matrix inversion operation and the eigenvalue decomposition for variable diagonal loading in the conventional LCMV beamforming, which avails the efficiency of beamforming, and the required operations are significantly reduced. The simulation results show that the improved methods own better recognition ability for the target signals and suppression ability for the interference signals.

In order to ensure the safe operation of special airports, RNP AR, a new navigation specification, is widely used. To meet the performance requirements of RNP AR for navigation systems, civil aircraft are generally equipped with multiple navigation sensors. Therefore, designing the integrated navigation algorithm and carrying out efficient fault-tolerant fusion processing of sensor information is the primary challenge in implementing RNP AR. In this paper, the mechanism of the influence of sensor anomalies on the integrated navigation system performance was explored, followed by the introduction of a new adaptive federated Kalman filter (FKF) structure. Simulation findings indicate that the accuracy performance of this method is superior to the traditional FKF, which provides a new idea to make sure that the civil aircraft meets the performance requirements of RNP AR.

In order to improve the measurement lower limit and anti-interference ability of the swirlmeter, the internal voltage-variable crystal acquisition circuit of the swirlmeter is given. A detection circuit composed of three Instrumentation amplifier and a voltage-parallel negative feedback circuit is adopted here to eliminate the effect of power-frequency noise interference, the input impedance is increased and the lower limit of the range is expanded, The experimental results show that the swirlmeter designed by the improved circuit has a wide range and anti-jamming performance, the accuracy can reach 1% in the measurement range.

A set of metal conductivity eddy current detection system is designed by using tunnel magnetoresistance (TMR) sensor and quadrature demodulation principle. The system uses one coil for excitation and dual TMR for receiving the magnetic field strength signals to differentiate. Through the finite element simulation, it is found that the logarithm of the phase of the difference between the magnetic field strengths at two points on the coil axis has a linear relationship with the logarithmic value of the metal conductivity. Therefore, in order to obtain the phase of the magnetic field, a quadrature demodulation system based on LabView is designed. The relative error of demodulation of magnetic field amplitude and phase of the system is within 1%. By measuring the electrical conductivity of the metal test blocks, we found that the experimental results are in good agreement with the simulation, and the relative errors of the conductivity are within 4%, which proves the feasibility of the system on detecting non-ferromagnetic materials.

To achieve good RF power detection performance, the thermocouple length of thermoelectric RF power sensor is studied in this article. The operation principle of the designed RF power sensor is firstly to convert the RF power into thermal power by terminal RF resistors and then convert the thermal power into DC voltage through many thermocouples based on Seebeck effect. Four the thermoelectric RF power sensors with thermocouple lengths of 200, 300, 400 and 500 μm are fabricated by the standard 0.18-μm CMOS technology. The experimental results demonstrate that measured reflection losses are all lower than -20dB in the operational frequency range of 8GHz to 12GHz. This indicates that the designed RF power sensors have good impedance matching characteristics in X-band. As the thermocouple length gradually increases, the measured sensitivities are 1.21 μV/mW, 1.48 μV/mW and 1.59 μV/mW and 1.61 μV/mW at the center frequency of 10 GHz. And the corresponding signal-to-noise ratios (SNR) are approximately 1.116 × 10⁵/W, 1.114 × 10⁵/W, 1.037 × 10⁵/W and 0.933 × 10⁵/W, respectively. The measurement results show that the increasing of the thermocouple length is helpful to improve the sensitivity of the sensor, but the corresponding signal-to-noise ratio will decrease. The results will help improve the future design of the thermoelectric RF power sensors thermocouple-based.

A new file detection system based on FPGA and TDLAS technology is proposed to solves the shortcomings of traditional fire detection sensors in airplane cabins, such as poor reaction time, high false alarm rate, and limited sensitivity. First, an FPGA-centered system architecture is constructed using gas molecule spectrum absorption theory. The FPGA is used to create a laser wavelength modulation signal generating module, as well as a high-speed data transmission module, a FIR digital low-pass filter module, and a digital phase-sensitive detector module. The system was used to evaluate the strength of gas absorption peaks at different concentrations. The second harmonic was extracted from the data while measuring and removing the interference from the external environment. The relative error of the final fitting results was less than 0.04%, which proves the feasibility of the design.

Cells play a very important role in biological experiments. To effectively detect cell activity, we fabricated a biomicroelectrode chip for cell activity detection using a semiconductor process, and then completed the package of a cell activity sensor by bonding the bio-microelectrode chip to a PCB using wire bonding technology and completing the encapsulation of the cell culture chamber on the bio-microelectrode chip. To verify the feasibility of the sensor for cell activity detection, CHO cells were grown in the cell culture cavity and cultured for 24 h. TrypLE (an excellent substitute for trypsin) was used to dissociate the adherent CHO and simulate the death of CHO cells, while the output impedance of the sensor was measured. The experimental results showed that the output impedance of the sensor changed significantly at 10^3~10^5 Hz during the gradual dissociation of CHO cells. Combined with the model analysis, it can be tentatively concluded that the sensor has the function of detecting cell activity.

The rapid recognition of flammable and toxic gases is an essential and challenging task for electronic noses (E-nose) in various fields. The traditional recognition method is limited by the limited features of the gas ultra-time response curve, and has the problems of unsatisfactory classification results and long recognition time. We propose a noval gas recognition algorithm that combines the ensemble learning method with kernel-base learning system learning (KBLS), thereby improving the accuracy of gas recognition, with faster recognition time, and a more stable model. We used four volatile combustible gases as gas test datasets and evaluated the algorithm proposed in this study over a range of response curves from 0.5s to 4s. According to our extensive comparative experiments, it is found that the accuracy is higher than that of other algorithms.

Railroad engineering geological survey is a general and important work in the railroad design process. The effective combination of remote sensing and GIS technology has broad application prospects in railroad survey, planning and construction. In this paper, we propose an image classification algorithm in railroad engineering geological survey, and determine the best segmentation scale through several experiments. Based on the spectral and geometric information features of different lithologies, a classification rule set with the KNN and SVM methods are used for lithology classification, and the accuracy evaluation shows that the classification results are more reliable. This work will greatly improve the segmentation efficiency and make the multi-scale segmentation technology of remote sensing images truly automated.

In order to avoid the independent on-line monitoring devices of transformers and the lack of unified data interfaces for various sensing devices, a broadband current array sensing system is proposed. The system includes high frequency current, core grounding current, power frequency current, over-voltage current and other sensing networks. The system is a causal linear time invariant system, based on stable control principle, the improved adaptive fault estimation algorithm is proposed, and system stability is calculated based on the system dynamics model and transfer function.

The catalytic activity of the oxygen-pumping electrode in automotive nitrogen oxide sensors has an essential effect on the final measurement accuracy of nitrogen oxides. To investigate the variation pattern of catalytic activity of Pt-Au alloy oxygen-pumping electrode affected by Au content and aging time, the corresponding three samples with various Au content were prepared by high-temperature co-fired ceramic (HTCC) process. All samples were subjected to long-term and high-temperature aging in a simulated automotive exhaust atmosphere, and the electrodes at different aging stages were tested for AC impedance and characterized for morphology. The results of testing show that the catalytic activity of the Pt-Au alloy electrodes decreases with the increase of Au content. At the same time, the catalytic activity of the electrodes also decays exponentially with the increasing aging time and stabilizes at about three weeks. The explanation is that as the Au content increases and aging proceeds, the electrochemical reaction triple phase boundaries (TPBs) decrease, and the catalytic activity of the electrodes decreases.

In the process of power transmission, changes in system parameters, such as inductance, capacitance and system operating frequency, will cause system detuning and reduce transmission performance. In this paper, a combined control system with frequency tracking is designed for the detuning of MCRWPT system caused by the parameter changes in the non overcoupling region. Only the phase controlled capacitor is added at the transmitter to tune, and at the same time, the phase-locked loop is used to track the frequency, so that the system can quickly recover the resonant state after detuning. This method does not need to add a circuit at the receiving end, which ensures that the system can obtain the best transmission characteristics while reducing the volume of electrical equipment.

This paper is based on GIS technology, through access to multi-source meteorological data, meteorological information, geographic information, and power grid equipment combined with correlation analysis, thus designed the power grid meteorological disaster warning information release system, realize the state grid disaster warning, statistical analysis, model, archives, professional maintenance, industry maintenance part of six power grid disaster warning information monitoring. The system can display all kinds of power disaster data information graphically, and realize the forecast and early warning of power disasters. The field experiment effect shows that the power grid meteorological disaster warning information system can provide professional meteorological warning support for realizing the fine management of the power grid and improving the disaster prevention capacity of the power grid.

In recent years, researchers have used various methods to study flexible sensors based on carbon nanomaterials, among which impedance analysis has unique advantages in studying the sensor mechanism. In this paper, we investigated the impedance characteristics of a flexible zero-dimensional carbon nanocomposite strain sensor, establishing a simple electrical model by electrochemical impedance spectroscopy (EIS). Based on the model, the sensing mechanism of the flexible strain sensor was analyzed. The noise analysis theory of the flexible strain sensor was established, and it is proved that the noise of the sensor can be reduced significantly when an AC signal is applied, which will improve the accuracy of the sensor.

Visual-inertial navigation system (VINS) is subject to degradation in urban vehicles. Specific motion patterns of the car and dynamic objects considerably reduce the pose estimation accuracy of VINS. A tightly-coupled visual-inertial-wheel odometry based on factor graph optimization with sliding window for ground vehicles is proposed in this paper. Wheel encoder and non-holonomic constraint information are utilized in pre-integration as an additional constraint to assist scale estimation. To promote the robustness of the system in complex environments, the prior pose from dead reckoning is employed to assist feature tracking and dynamic object filtering. The results from the real-world experiments in large-scale urban driving environments show better robustness and accuracy compared to the state-of-the-art methods.

Most of the existing hyperspectral imagers can provide the image of insulators. Traditional optical elements such as prism and grating can only reflect the surface pollution in a certain period of time. In this paper, a new type of splitter, linear gradation filter, is proposed to replace the former grating and prism for splitting light. Compared with the traditional hyperspectral imaging system using prism or grating spectroscopy, the filter based hyperspectral camera has smaller volume, wider scanning range and higher spatial and spectral resolution.

Risk identification of single-person field operation can ensure the safety of power grid patrol personnel. In order to accurately identify the risk of single-person field operation, a dynamic risk identification method of single-person field operation combining infrared living induction and ultrasonic technology is proposed. According to the principle of infrared living induction monitoring, the relationship between the input value and the output value of infrared living induction equipment is established. The single-person field operation risk is monitored by using the spectral response function curve. By calculating the risk characteristic parameters of single-person field operation, the correlation coefficient of each risk characteristic is obtained. According to the dynamic change parameters of threshold, the single-person field operation risk characteristics are extracted. By setting the optimal distance between field operators and ultrasonic detection devices, the optimal correlation characteristics of single-person field operation risk are calculated. By applying ultrasonic technology, the objective function of single-person field operation risk identification is constructed to identify single-person. The experimental results show that the method in this paper has a good driving effect on wild animals under different lengths of ultrasonic waves, and can accurately identify the risk of single-person field operation, meeting the safety requirements of power grid patrol personnel.

Inertial measurement unit (IMU) has been commonly used in measuring angles, displacement measurement and other fields. However, micro electro-mechanical system (MEMS) gyroscopes have little accuracy of measuring and enormous noise; real signals will be cloaked in the noise. A new denoising algorithm for low cost MEMS gyroscope is offeredin this paper. First, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm is used to decompose the initial-signal which can obtain a limited number of intrinsic mode functions (IMF). Next, use the power spectrum entropy to divide the IMF into noise signals, mixed signals and sound signals. Then, variational mode decomposition (VMD) is choosen to decompose the mixed signal, and using reptile search algorithm (RSA) optimization algorithm optimizes the crucial arguments of VMD. VMD algorithm appropriately dismantles the signal according to the optimized arguments to attain the IMF sections, divided into signal sections and noise sections using the Bhattacharyian distance. The signal sections are furthermore processed by lifting wavelet threshold denoising (LWT), and the valuable signal with denoising is obtained. Finally, all the valuable signals are amalgamated and reestablished into the ending-denoising signal. The effectuality of the algorithm is weighted by Matlab simulation. The static denoising experimentation of the three-axis turntable is put up in the experimental analysis, which exhaustively confirms that the algorithm has clear superiority in denoising and significantly increases the signal quality and the accuracy of the cheap MEMS gyroscope.

To guarantee the working ability and safety of the airport single light monitoring system, the single light monitoring information transmission method based on the combination method of power-line carrier and wireless sensor network is proposed. This method combines the advantages of both, adopts wireless detection node monitoring data, the center theory algorithm of graph theory and Dijkstra algorithm to select cluster head and design the best route, and transmits the error information to the sink node; the narrowband power-line carrier technology modulated the information to the power line to achieve long-distance stable transmission. The combination of these two methods aim at reducing the network energy consumption of the wireless sensor network, balancing the distribution of energy consumption, extending network life cycle, improving the transmission performance of power-line carrier technology signal, complementing the defects of the two, and effectively ensuring the safety of information transmission.

Aiming at the problem that discrete emotion recognition cannot depict continuous emotion changes, in order to capture high-level dimensional emotional information, this paper integrates attention mechanism into the two stream CNN model and proposes a Two Stream Convolutional Neural Network with Shared and Global attention mechanism (TSCNN-SGA) .TSCNN-SGA uses the same structure of CNN network structure to extract the static stream of expression images and dynamic stream of expression sequences features respectively, firstly, in the dynamic and static dual flow feature extraction network, the output feature map of the previous convolution layer group is used to cascade to calculate the shared attention weight of the next layer group, secondly, the two stream convolution feature map with shared attention is cascaded, the attention weights of different positions are mapped onto the cascaded feature map and weighted, finally, the shared weight matrix in the convolution end of TSCNN-SSA and the global attention mechanism after the two stream feature cascade work together to obtain the depth space-time feature, which is input to the bidirectional long-short time network to obtain the final dimensional sentiment prediction value. Compared with different baseline methods, the average value of the proposed method's concordance correlation coefficient (CCC) in the arousal-valence space reached 0.576, which can effectively identify dimensional emotions.

Inertial navigation is widely used in many fields, such as land, underwater and aviation. Aiming at the shortcoming that the cumulative error of the strapdown inertial navigation system is difficult to be effectively corrected during long-term navigation, it cannot meet the accuracy requirements in the long-term navigation process. In this paper, based on the strapdown inertial navigation, according to the global distribution of the geomagnetic field and the unique characteristics of the magnetic field strength, a second-order complementary filter strapdown inertial navigation algorithm is proposed to overcome the shortcoming that a single inertial navigation system cannot satisfy long-term precise navigation. And the improved algorithm is simulated and analyzed.

A neural network prediction model for NMOFET device is proposed in this paper by using BP algorithm in machine learning technology and Silvaco TCAD simulation tool, so as to improve the efficiency of actual simulation work and provide a new method for the performance research of NMOSFET. In the process of modeling, the substrate bias, the impurity concentration of substrate, the thickness of oxide and the threshold voltage adjustment implant doping concentration are regarded as independent variable, and substituted into the simulation software for simulation. The performance parameters of NMOSFET are obtained as the dependent variable, such as the threshold voltage, the maximum transconductance, the subthreshold slope and the I_on/I_off ratio. All the simulation results are used to assisted the training and prediction of BP neural network. The results of experiment show that the BP neural network prediction model with Levenberg-Marquardt algorithm can well predict the performance of NMOSFET devices. The determination coefficient of all models is above 0.99, and the mean absolute percentage error of all models is below 1.5%.

During the operation of the substation, the main reason for the heating of the switchgear is the eddy current loss. In the daily operation and maintenance of the substation and loss statistics, for some switchgears in the area of direct conductive equipment, there is an abnormal local overheating phenomenon. The main reason for this phenomenon is the eddy current loss in the substation, which seriously affects the safe operation of substation equipment due to the heating defect. This paper analyzes the local overheating problem in substation switchgear and proposes the distribution of eddy current loss and its influencing factors of optimization strategy from the aspect of loss source design. Based on the Maxwell equation, the loss calculation model is established. The optimization method of eddy current loss is comprehensively evaluated according to the distribution characteristics of eddy current loss. The temperature of the bushing surface under different voltage levels is measured, and the error between the calculated results and the measured results is within 2%. Therefore, within the allowable error range, the simulation results and the formula calculation results can fit the measured values well, which verifies the correctness of the established wall bushing heating simulation model and temperature calculation formula. The optimal solution is made to provide a reliable basis for the actual engineering construction. The electric field energy of the bushing is relatively evenly distributed under the actual operating conditions, and the insulating materials are fully utilized to ensure the reliability under high ambient temperature and high current operating conditions.

For the characteristics of high mobility and dynamic topology of UAVs, it is important to select a suitable mobility model for DTN’s performance. In this scenario, the Shortest Path Map Based (SPMB) Movement based on the UAV DTN is proposed. The background of DTN and the motion modes of RWP, GM and SPMB are described. In ONE simulator, controlling other variables unchanged, and the above three mobile models are used to simulate the UAV network nodes in the Spray-and-Wait routing algorithm. The results indicate that compared with RWP and GM, the message delivery ratio of SPMB is increased by 12% and 7.6% respectively. The network overhead ratio decreased by 7.27 % and 4.44%, respectively. The average delay is reduced by 798.38s and 564.43s respectively. In conclusion, the DTN routing protocol of SPMB is more reliable in the scenario of UAVs.

After preparing an all-solid-state copper ion selective electrode (Cu²⁺-ISE) based on PEDOT/PSS conversion layer via electrodeposition, this paper characterizes the morphology and cyclic voltammetry performance of the PEDOT/PSS film, optimizes the deposition time of the PEDOT/PSS film, and tests the reproducibility, selectivity and acid-base resistance of the all-solid-state copper ion selective electrode. The experimental results show that the prepared all-solid-state copper ion selective electrode has a linear response in the range of 1×10^-6 mol·L^-1~1×10^-2 mol·L^-1, with a response time less than 30s, and the response slope of 27.8 mV/ decade. The all-solid-state copper ion selective electrode mentioned above can meet the detection standard of national II surface water.

Indoor geolocation technique based on the time-of-arrival (TOA) has advantages of low cost, high accuracy compared with other location technologies. However, traditional TOA estimation methods based on signal correlation or peak finding have difficulty keeping high ranging accuracy and robustness in multipath conditions. To address the problem, an efficient and easily-implemented algorithm as a high-resolution TOA estimator in multipath environments is presented in this paper. Numerical simulation is conducted to verify the performance of the proposed TOA estimation algorithm. Results show the algorithm outperforms conventional method in estimation accuracy and robustness.

In order to improve the accuracy of the Off-Block Time and incorporate uncertainties in the prediction process. Firstly, this paper considers the key influencing factors of AGS (Aviation Ground Support) and simplifies the complex Turn-Round process. Secondly, a method for estimating the time based on LSTM-TCN hybrid network model is proposed, and the uncertainty that may occur in the prediction process is quantified by using MC Dropout approximate Bayesian neural network. Select a single flight departure operation data of an airport in central China to verify. The results show that the accuracy of each node interval can reach 80 %, and the reliability is more than 90 %.

This work develops a solid-state nanopore sensing system for the detection of biological nanoparticles. A silicon-based silicon nitride solid-state nanopore chip with a low aspect ratio was designed and made by the microfabrication technique. As a demonstration, the chip was used to characterize and analyze carboxyl-modified polystyrene particles. Time series analysis was used to construct the characteristics of the current signal during nanoparticle translocation. Time series analysis based on machine learning algorithms combined with Bayesian optimization was used to perform signal discrimination of particles of different sizes. This sensing technique has promise for identifying the characteristics of nanoparticles like proteins, viruses.

In recent years, deep learning has been widely studied in soft sensor modeling. However, the prediction of the deep learning model is difficult to explain, and it is hard to incorporate prior experience into the model. These shortcomings of deep learning prevent its application in real industrial processes. In this article, we propose a self-adaptive graph convolution networks (SAGCN) for industrial soft sensor modeling. This model uses the graph convolution network to introduce prior knowledge and construct the displayed nonlinear relationship among variables. And the graph convolution network can aggregate information to extract features from data. Because it is difficult and highly subjective to rely on prior knowledge and mechanisms to obtain the graph structure, this article proposes a graph structure self-learning method to realize the joint learning of the nonlinear relationship among auxiliary variables and the regression relationship between auxiliary variables and quality variables. The proposed method is verified through the CO₂ absorption column process from a real ammonia synthesis process. Based on the results, SAGCN demonstrates high accuracy and a certain capacity to discover knowledge.

Aiming at the complexity and difficulty of millimeter wave radar target detection algorithm, SYS/BIOS is used to design and implement the algorithm. Due to the large amount of computation and tasks, the algorithm of millimeter wave radar target detection is implemented by SYS/BIOS embedded operating system, rather than developing based on bare computer. The operating system is modular, so it is convenient to manage and develop programs, and effectively solves the difficult problem of millimeter wave radar target detection algorithm. The feasibility of the algorithm is proved by practical experiment.

The traditional mobile phone signal blocker adopts high power broadband noise suppression technology, which has poor jamming effect and has adverse effects on the health of people in the shielded area. This paper proposes a synchronous coherent jamming technology that jams with 5G mobile phone signals. According to the characteristics of 5G base station downlink signals, the technology can conduct coherent jamming modulation to the synchronous signal in the process of mobile phone search and destroy the cell search process of mobile phone terminals. It is suitable for 5G, 4G, and other mobile phone terminals. The simulation results show that the jamming signal is coherent with the base station downlink signal, and has the advantages of narrow jamming band width and high power utilization.

Fiber optic gyro (FOG) is a kind of angular velocity gyro based on the Sagnac effect, and the high reliability of long lifespan of the FOG are very important to the development of high requirement security control equipment. Spherical inertial measurement unit is designed, six axis redundancy FOGs is adopted, and the sensitive axis of any two gyros are not colinear, and the sensitive axis of any three gyros are not coplanar in this combination structure with a highly reliable orthogonal oblique configuration, when one of the sensitive components fails, other FOGs can also ensure the normal operation of the system, which greatly improves the accuracy, reliability and stability of the gyroscope. In the inertial measurement unit (IMU), the system works properly as long as the gyros in three different directions are working properly.

To provide people with a safer and more comfortable living environment, a smoke temperature and humidity system based on an AT89C52 microcontroller is designed, which can use an MQ-2 smoke sensor to collect toxic and harmful gases in the air, and then transmit the data to the microcontroller through AD conversion, and also design a temperature and humidity monitoring module, which can be sent to the microcontroller through the DHT11 sensor sampling into digital quantities. Finally, the smoke concentration collected by the two sensors as well as the temperature and humidity values are displayed through the LCD, and an alarm feedback information module is also designed. The design is simple to operate and has good stability.