Search

Analysis of sign language characteristics is of great significance to the establishment of sign language recognition systems. However, a unified standard is required to evaluate whether the collected and analyzed characteristics are complete and can effectively characterize sign language gestures. In this study, a complete upper limb kinematics model is established for sign language based on the Denavit–Hartenberg (DH) robotics kinematics model; the expression of key nodes and the determination method of model parameters based on upper limb sign language are analyzed; the inverse kinematics solving equation is established according to the actual sign language expression, and the correctness of the platform inverse kinematics equation is verified. This study establishes a theoretical basis for the application research of the platform so as to provide model support for judging the completeness of the characteristic analysis.

Nonnegative matrix factorization (NMF) is a powerful tool for parameter estimation applied in numerous robotics applications, such as path planning, motion trajectory prediction, and motion intention detection. In particular, NMF has been successfully used to extract simplified and organized movement primitives from myoelectric signal (MES) for robust control of multi-degree of freedom humanoid robots. However, MES is typically contaminated by complex noise sources. The system performance often degrades due to the simplified Gaussian assumption of the noise distribution in existing NMF methods. Furthermore, most existing NMF models are unable to automatically determine the rank of the latent matrices. To address these issues, this article presents a hybrid variational Bayesian Gaussian mixture and NMF (GMNMF) model with a finite Gaussian mixture model adopted to fit the mixed noise density function of MES. In addition, the automatic relevant determination criterion is applied to automatically infer the number of movement primitives. The coordinate descent update rules for the proposed model are formulated by mean-field variational Bayesian inference. We assess the model performance on five synthetic noise distribution functions and an experimental MES dataset to perform six wrist movements. The results demonstrate that GMNMF yields low error and high robustness in extracting the movement primitives over four competitive methods for robust cybernetic control.

The long-distance acoustic signal has low energy, high concealment and low signal-to-noise ratio (SNR), thus is often submerged in noise and interference. The conventional method of acoustic detection cannot classify or recognize the target. A sound recognition method has been proposed to simulate human ear auditory process by making the best of sound recognition advantage of human ear. First of all, those parts which function as energy gathering, frequency-selective amplification or information transmission in human ear are selected, and the auditory model of each part is established independently. Then, the structural parameters of each model are extracted, and the influence and change rules of structural parameters on the auditory signal are studied and optimized according to the auditory law. Finally, the auditory model is reconstructed based on the parameters extracted from each component, and the integrated model is used to carry out the voice recognition experiment under low SNR. The recognition rate of low SNR (−5 dB) acoustic event is improved by 6.5%.

The principle and method of existing indoor shadow photography system and shadow target calibration system are introduced. The effects of muzzle shock wave, projectile shock wave, trajectory descent of projectile flight and natural wind, dust, ground irregularity and stray light in the field measurement environment when using Shadow Photography to measure the Motion Attitude of Large Caliber Projectile are analyzed. The key parameters of the large caliber projectile motion attitude measurement system in the field are determined, and the key links that need to be paid attention to in the design are put forward accordingly.

In order to effectively detect the fatigue status of equipment operators, this paper proposes a fatigue detection model that integrates multi-source physiological signals to evaluate the fatigue level of equipment operators in order to improve the efficiency of military operations and training. This method selects various physiological signals such as human heart rate, heart rate variability, body temperature, respiratory rate, blood oxygen concentration, etc. according to the working environment and working characteristics of the equipment operator, and establishes the relationship between the characteristic values of various physiological parameters and fatigue levels, And then use the optimized support vector machine to perform fusion judgment on the feature values of multi-source physiological parameters, and obtain the fatigue state of the equipment operator. Experimental results show that this method can make a more accurate judgment on the fatigue state of equipment operators, and is of great significance for ensuring the efficiency of equipment operators’ work and training.

Surrounded by low SNR, how to make the voice faster and better recognize the owner has become a heated research topic. The human auditory system can accurately acquire the characteristics of acoustic events in complex systems or low SNR noise environment, which is of significance in the research of bionic hearing of human ear. The response curve of human ear output is obtained by bionic technology, which is the best response curve for sound enhancement to modify Mel filter. The method of adaptive threshold selection is used to integrate Mel features to realize the reduction and dynamic extraction of low SNR speech features. This method not only can resist the disadvantages of poor robustness and complexity of parameter model, but also obtain dynamic and comprehensive speech information of different speakers in different scenes. Finally, the improved CNN and I-vector system are contributed to reduce the dimension of the data and to verify the recognition, so as to achieve the optimal frequency selective amplification and simplification of the acoustic signal. In the case of SNR-5db, the model is reduced by 15% and the recognition accuracy is improved by 3%.

To meet the large-caliber projectile motion measurement system calibration requirement, developed a large-caliber projectile motion measurement system calibration system, gives the calibration principle and working process, design a visibility good green light source as the collimating light source, the design targets based on the filtering and PSD field grid position detection module; Aiming at the calibration requirement that the six faces of cavity cross grids frame cannot be shielded, the cavity cross grids adjustment module is designed. Error analysis shows that the calibration error of the system is 0.062mm.

According to the shape characteristics of large-caliber projectiles in shadow photography and the characteristics of external field environment, this paper focuses on the characteristics of external field shadow imaging and proposes a method for extracting the features of outfield shadow image of large-caliber projectiles. Based on the image preprocessing, the Bullet Tip feature points are determined by the intersection of the central axis and the edge, while the bullet tail feature extraction method based on ellipse fitting. The fitting accuracy of ellipse fitting method and upper and lower endpoint method are simulated and compared. The results show that the ellipse fitting method determining the center of the target end surface is better, with the extraction accuracy of 0.44 piexl.

The calibration principle and working process of shadowgraph station for large caliber projectile motion attitude measurement is studied. The main error factors affecting the calibration accuracy of the system are analysed, and the various errors in the calibration process are statistically analysed. The calibration error of the measurement system is calculated to be 0.062 mm. The error analysis process shows that the coincidence error between the laser optical axis and the central axis of double square nets is the main error of calibration. Among these errors, the installation error of optical window and the centering error of filtering target (Angle deviation between parallel light source beam axis and double square nets axis caused by coincidence error of reflected facula and incident facula) are the main sources of errors.

Aimed at obvious interference noise of shadow images of bullet shot in outside field, and larger extraction error of bullet attitude features of shadow images and other problems, intersection point location method based on axis line and edge is adopted to extract features of bullet tip, and the method based on ellipse fitting is adopted to extract features of bullet tail in the Paper, and then extracted feature information is used to calculate motion attitude of bullets. The method can be applied to shadow image analysis of bullets under field environment well with accurate feature extraction, good consistency and high analysis precision and other features.

In order to increase electric cars power efficiency when control frequency conversion, and reduce the interference of high harmonics, we propose a design method of voltage space vector (SVPWM) frequency conversion control module. Based on TMS320LF2407 the PWM signal producing module design, IPM module as the core, we design the inverter circuit and Developed SVPWM real-time motion control program. This program use the manner of segmentation modulation, achieve frequency conversion control a range of 5 ~ 100Hz frequency. Variable frequency speed-governing experiment shows that, The control method can reduce the switching losses, meanwhile suppress the temperature rise.