LiDAR DNN based self-attitude estimation with learning landscape regularities

This paper presents an EKF (extended Kalman filter) based self-attitude estimation method with a LiDAR DNN (deep neural network) learning landscape regularities. The proposed DNN infers the gravity direction from LiDAR data. The point cloud obtained with the LiDAR is transformed to a depth image to be input to the network. It is pre-trained with large synthetic datasets. They are collected in a flight simulator because various gravity vectors can be easily obtained, although this study focuses not only on UAVs. Fine-tuning with datasets collected with real sensors is done after the pre-training. Data augmentation is processed during the training in order to provide higher general versatility. The proposed method integrates angular rates from a gyroscope and the DNN outputs in an EKF. Static validations are performed to show the DNN can infer the gravity direction. Dynamic validations are performed to show the DNN can be used in real-time estimation. Some conventional methods are implemented for comparison.

Understanding Vestibular Information Processing in Velocity-Storage Circuit and Application to Interpreting Clinical Manifestation of Vestibular Disorders

The velocity-storage circuit comprised of bilateral vestibular nucleus complexes, commissural fiber, and nodulus and uvula functions in refining the raw vestibular signal to estimate rotational velocity, gravity direction, and inertia. In this review, we pursued the functional significance of this velocity-storage circuit and how this physiologic knowledge could help us understand the clinical symptoms and signs of patients with vestibular disorders.

Study on Deposition Characteristics of Microparticles in Terminal Pulmonary Acini by IB–LBM

As an indicator of health risk, the deposition of microparticles in terminal pulmonary acini is of great significance in the medical field. To control particulate pollution and optimize aerosol delivery, it is necessary to perform an in-depth study of the microparticle deposition in terminal pulmonary acini; however, little research has been done on this topic. This paper proposes a respiratory movement model of terminal pulmonary acini using an immersed boundary–lattice Boltzmann method. In addition, we explored the effect of gravity direction, respiratory rate, microparticle diameter, and other parameters on the microparticles deposition process and distribution, under the airflow in the acinar wall. It was found that the deposition of microparticles is sensitive to gravity direction, and the growth of the respiratory rate increases the rate of microparticle migration and deposition. It was observed that the gravity effect is enhanced by increasing the diameter of microparticles, causing a high deposition and dispersion rate. The study reveals the dynamic correlation between the respiration process and the movement of microparticles, which is of reference value to figure out the pathogenicity mechanism of inhalable particles and to optimize the aerosol delivery.

Fast Monocular Visual-Inertial Initialization with an Improved Iterative Strategy

The initialization process has a great effect on the performance of the monocular visual inertial simultaneous localization and mapping (VI-SLAM) system. The initial estimation is usually solved by least squares such as the Gauss-Newton (G-N) algorithm, but the large iteration increment might lead to the slow convergence or even divergence. In order to solve this problem, an improved iterative strategy for initial estimation is proposed. The methodology of our initialization can be divided into four steps: Firstly, the pure visual ORB-SLAM model is utilized to make all variables observable. Secondly, the IMU preintegration technology is adopted for IMU-camera frequency alignment at the same time with key frame generation. Thirdly, an improved iterative strategy which is based on the trust region is introduced for the gyroscope bias estimation as well as the gravity direction is refined. Finally, the accelerometer bias and visual scale are estimated on the basis of previous estimations. Experimental results on the public datasets show that the estimation of initial values can be converged faster, as well as the velocity and pose of sensor suite can be estimated more accurately than the original method.

Transient simulation on dynamic response of liquid annular seals

Transient change of the operating parameters has a serious influence on the stability of liquid annular seals. Take the liquid annular seals as a research object, a numerical method based on six-degree-of-freedom (6DOF) to analyze the dynamic response of liquid annular seals under gravity impact load. The variations of the force of liquid seal and pressure as well as the axis trajectory in time history are investigated. The influence of different sealing clearance、different liquid viscosity and different rotor speed is also studied. The results show that the maximum sealing pressure and sealing force of gravity direction will increase greatly in a very short time and then reduce rapidly. When sealing clearance increases, the displacement response amplitudes of axis trajectory, the maximum sealing force of gravity direction and maximum sealing pressure also increase. When liquid viscosity increases, the displacement response amplitudes of axis trajectory, the maximum sealing force of gravity direction and maximum sealing pressure decrease. We also found that different rotor speed has almost no influence on the maximum sealing force of gravity direction and maximum sealing pressure.