Search

The registration of laser point clouds in complex conditions in wilderness scenes is an important aspect in the research field of autonomous vehicle navigation. It serves as the foundation for solving problems such as environment reconstruction, map construction, navigation and positioning, and pose estimation during the motion process of autonomous vehicles using laser radar sensors. Due to the sparse structured features, uneven point cloud density, and high noise levels in wilderness scenes, achieving reliable and accurate point cloud registration is challenging. In this paper, we propose a semantic-supervised sparse point cloud registration network (S3PCRNet) aiming to achieve effective registration of laser point clouds in wilderness large-scale scenes. Firstly, a local feature aggregation module is designed to extract the local structural features of the point cloud. Then, based on rotation position encoding, a randomly grouped self-attention mechanism is proposed to obtain the global features of the point cloud through learning. A semantic information weight matrix is calculated to filter out negligible points. Subsequently, a semantic fusion feature module is utilised to find reliable correspondences between point clouds. Finally, the proposed method is trained and evaluated on both the RELLIS-3D dataset and a self-made Off-road-3D dataset.

Abstract Coronavirus disease 2019 (COVID-19) has emerged, rapidly spread and caused significant morbidity and mortality worldwide. There is an urgent public health need for rapid, sensitive, specific, and on-site diagnostic tests for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. In this study, a fully integrated and portable analyzer was developed to detect SARS-CoV-2 from swab samples based on solid-phase nucleic acid extraction and reverse transcription loop-mediated isothermal amplification (RT-LAMP). The swab can be directly inserted into a cassette for multiplexed detection of respiratory pathogens without pre-preparation. The overall detection process, including swab rinsing, magnetic bead-based nucleic acid extraction, and 8-plex real-time RT-LAMP, can be automatically performed in the cassette within 80 min. The functionality of the cassette was validated by detecting the presence of a SARS-CoV-2 pseudovirus and three other respiratory pathogens, i.e., Klebsiella pneumoniae, Pseudomonas aeruginosa, and Stenotrophomonas maltophilia. The limit of detection (LoD) for the SARS-CoV-2 pseudovirus was 2.5 copies/μL with both primer sets (N gene and ORF1ab gene), and the three bacterial species were successfully detected with an LoD of 2.5 colony-forming units (CFU)/μL in 800 μL of swab rinse. Thus, the analyzer developed in this study has the potential to rapidly detect SARS-CoV-2 and other respiratory pathogens on site in a “raw-sample-in and answer-out” manner.

Reliable detection of specific antibodies against pathogens by lateral flow immunoassay (LFIA) greatly depends on the composition of the detectable complex and the order of its assembly. We compared three LFIA formats for revealing anti-SARS-CoV-2 antibodies in sera with the following detected complexes in the analytical zone of the strip: antigen–antibodies–labeled immunoglobulin-binding protein (Scheme A); antigen–antibodies–labeled antigen (Scheme B); and immunoglobulin-binding protein–antibodies–labeled antigen (Scheme C). The lowest detection limit was observed for Scheme C, and was equal to 10 ng/mL of specific humanized monoclonal antibodies. When working with pooled positive sera, Scheme C had a detection limit 15 times lower than Scheme B and 255 times lower than Scheme A. Due to the high sensitivity of Scheme C, its application for the panel of human sera (n = 22) demonstrated 100% diagnostic specificity and sensitivity. These consistent results be useful for designing the format of LFIA serodiagnosis for other diseases.

This paper proposes a clothoid-curve-based trajectory tracking control method for autonomous vehicles to solve the problem of tracking errors caused by the discontinuous curvature of the control curve calculated by the pure pursuit tracking algorithm. Firstly, based on the Ackerman steering model, the motion model is constructed for vehicle trajectory tracking, Then, the position of the vehicle after the communication delay of the control system is predicted as the starting point of the clothoid control curve, and the optimization interval of the curve end point is determined. The clothoid control curves are calculated, and their parameters are verified by the vehicle motion and safety constraints, so as to obtain the optimal clothoid control curve satisfying the constraints. Finally, considering the servo system response delay time of the steering system, the steering angle target control value is obtained by previewing the curvature of the clothoid control curve. The field experiment is conducted on the test road, which consists of straight, right-angle turns and lane-change elements under three sets of speed limitations, and the test results show that the proposed clothoid-curve-based trajectory tracking control method greatly improved the tracking accuracy compared with the pure pursuit method; in particular, the yaw deviation is improved by more than 50%.

Respiratory tract infections such as the ongoing coronavirus disease 2019 (COVID-19) has seriously threatened public health in the last decades. The experience of fighting against the epidemic highlights the importance of user-friendly and accessible point-of-care systems for nucleic acid (NA) detection. To realize low-cost and multiplexed point-of-care NA detection, a swing-assisted multiplexed analyzer for point-of-care respiratory tract infection testing (SMART) was proposed to detect multiple respiratory tract pathogens using visible loop-mediated isothermal amplification. By performing hand-swing movements to generate acceleration force to distribute samples into reaction chambers, the design of the SMART system was greatly simplified. By using different format of chips and integrating into a suitcase, this system can be applied to on-site multitarget and multi-sample testing. Three targets including the N and Orf genes of SARS-CoV-2 and the internal control were simultaneously analyzed (limit of detection: 2000 copies/mL for raw sample; 200 copies/mL for extracted sample). Twenty-three clinical samples with eight types of respiratory bacteria and twelve COVID-19 clinical samples were successfully detected. These results indicate that the SMART system has the potential to be further developed as a versatile tool in the diagnosis of respiratory tract infection.

Long Short-Term Memory (LSTM) neural network has been widely used in many applications, but its application in classification of vehicle movement patterns is still limited. In this paper, LSTM is applied to the vehicle behavior recognition problem to identify the left turn, right turn and straight behavior of the vehicle at the intersection. On the basis of the traditional LSTM classification model, this paper transversely merges the input features and then inputs into a LSTM cell to get an improved model. The improved model can make full use of the input information and reduce unnecessary calculations, and the output of a single LSTM cell model can filter out interference information and retain important information, so it has better classification effect and faster training speed. The experimental results show that the proposed improved LSTM network classification model in this paper has a significant improvement in recognition accuracy and training speed compared with the improved model, the accuracy is increased by 1.6%, and the training time is reduced by 3.96 s. In addition, this paper also applies the improved model to regression problems, emotion classification and handwritten digit recognition and all of them have a good improvement effect, which improves the applicability and stability of LSTM in classification problems and provides a new way to deal with classification problems.

With the increasing application of dual-PTZ (Pan-Tilt-Zoom) cameras in intelligent unmanned systems, research regarding their calibration methods is becoming more and more important. The intrinsic and extrinsic parameters of dual-PTZ cameras continuously change during rotation and zoom, resulting in difficulties in obtaining precise calibration. Here, we propose a general calibration method for dual-PTZ cameras with variable focal length and posture under the following conditions: the optical center of the camera does not coincide with the horizontal and pitch rotation axes, and the horizontal and pitch rotation axes are not perpendicular to each other. We establish a relationship between the intrinsic and extrinsic parameters and the feedback parameters (pan, tilt, zoom value) of dual-PTZ cameras by fitting and calculating previous calibration results acquired at specific angles and zoom values using Zhang’s calibration method. Subsequently, we derive the intrinsic and extrinsic parameter calculation formula at arbitrary focal length and posture based on the camera’s feedback parameters. The experimental results show that intrinsic and extrinsic parameters computed using the proposed method can better meet precision requirements compared with the ground truth calibrated using Zhang’s method. The average focal length error is less than 4%, the cosine similarity of the rotation matrix between the left and right cameras is more than 99.8%, the translation vector error is less than 1%, and the recalculated Euler angle errors are less than 1 degree. Our work can quickly and accurately obtain intrinsic and extrinsic parameters during the use of the dual-PTZ camera.

A numerical model of 84 high-strength concrete-filled square steel tubular columns (HSCFST) with local defects is developed using ABAQUS. The effects of parameters such as crack angle, crack length, impact surface and impact energy on the impact resistance of HSCFST columns are considered. The results show that under the effect of local corrosion, a model with horizontal cracks will show the phenomenon of crack closure when subjected to the front impact. The impact force platform value is mainly affected by the impact surface, followed by the crack angle, while the increase of the crack length mainly has a greater effect on the model of the rear impact. The impact resistance of the front impact model is better than that of the side and rear impact models. Increasing the crack length and decreasing the crack angle will increase the mid-span deflection of the model, and the mid-span deflection of the front impact model is smaller than that of the side and rear impact models. The energy absorption ratio of the model is proportional to the increase of the crack length and inversely proportional to the increase of the crack angle. Decreasing the crack angle will reduce the increase coefficient (Rd) of the dynamic flexural capacity of the model. A practical calculation method for the increased coefficient of the dynamic flexural capacity of HSCFST columns under local corrosion is proposed.

Topological maps generated in complex and irregular unknown environments are meaningful for autonomous robots’ navigation. To obtain the skeleton of the environment without obstacle polygon extraction and clustering, we propose a method to obtain high-quality topological maps using only pure Voronoi diagrams in three steps. Supported by Voronoi vertex’s property of the largest empty circle, the method updates the global topological map incrementally in both dynamic and static environments online. The incremental method can be adapted to any fundamental Voronoi diagram generator. We maintain the entire space by two graphs, the pruned Voronoi graph for incremental updates and the reduced approximated generalized Voronoi graph for routing planning requests. We present an extensive benchmark and real-world experiment, and our method completes the environment representation in both indoor and outdoor areas. The proposed method generates a compact topological map in both small- and large-scale scenarios, which is defined as the total length and vertices of topological maps. Additionally, our method has been shortened by several orders of magnitude in terms of the total length and consumes less than 30% of the average time cost compared to state-of-the-art methods.

The presence of pathogen-specific antibodies in the blood is widely controlled by a serodiagnostic technique based on the lateral flow immunoassay (LFIA). However, its common one-stage format with an antigen immobilized in the binding zone of a test strip and a nanodispersed label conjugated with immunoglobulin-binding proteins is associated with risks of very low analytical signals. In this study, the first stage of the immunochromatographic serodiagnosis was carried out in its traditional format using a conjugate of gold nanoparticles with staphylococcal immunoglobulin-binding protein A and an antigen immobilized on a working membrane. At the second stage, a labeled immunoglobulin-binding protein was added, which enhanced the coloration of the bound immune complexes. The use of two separated steps, binding of specific antibodies, and further coloration of the formed complexes, allowed for a significant reduction of the influence of non-specific immunoglobulins on the assay results. The proposed approach was applied for the serodiagnosis using a recombinant RBD protein of SARS-CoV-2. As a result, an increase in the intensity of test zone coloration by more than two orders of magnitude was demonstrated, which enabled the significant reduction of false-negative results. The diagnostic sensitivity of the LFIA was 62.5% for the common format and 100% for the enhanced format. Moreover, the diagnostic specificity of both variants was 100%.

In order to make an accurate prediction of vehicle trajectory in a dynamic environment, a Unidirectional and Bidirectional LSTM (UB-LSTM) vehicle trajectory prediction model combined with behavior recognition is proposed, and then an acceleration trajectory optimization algorithm is proposed. Firstly, the interactive information with the surrounding vehicles is obtained by calculation, then the vehicle behavior recognition model is established by using LSTM, and the vehicle information is input into the behavior recognition model to identify vehicle behavior. Then, the trajectory prediction model is established based on Unidirectional and Bidirectional LSTM, and the identified vehicle behavior and the input information of the behavior recognition model are input into the trajectory prediction model to predict the horizontal and vertical speed and coordinates of the vehicle in the next 3 seconds. Experiments are carried out with NGSIM data sets, and the experimental results show that the mean square error (MSE) between the predicted trajectory and the actual trajectory obtained by this method is 0.124, which is 97.2% lower than that of the method that does not consider vehicle behavior and directly predicts the trajectory. The test loss is 0.000497, which is 95.68% lower than that without considering vehicle behavior. The predicted trajectory is obviously optimized, closer to the actual trajectory, and the performance is more stable.

Abstract Although culture-based identification of bacteria is the gold-standard for the diagnosis of infectious diseases, it is time consuming. Recent advances in molecular diagnostics and microfluidic technologies have opened up new avenues for rapid detection of bacteria. Here, we describe a centrifugal-microfluidic chip for the detection of bacteria by integrating the cell lysis, clarification, and loop-mediated amplification (LAMP). The major advantages of this chip are as follows. Firstly, bacteria lysis was innovatively achieved by rotating a pair of magnets to generate bead-beating while the chip was kept stationary during lysis, which simplified the chip design because no additional valve was needed. Secondly, the on-chip assay time was short (within 70 min), which was competitive in emergency situations. Thirdly, results of the analysis can be interpreted by using a fluorescence detector or by the naked-eye, making it versatile in many areas, especially the resource-limited areas. The on-chip limits of detection of six types of bacteria were valued by gel electrophoresis, showing the similar results compared to the bench-top LAMP protocol. This chip can be used for rapid, sensitive, accurate and automated detection of bacteria, offering a promising alternative for simplifying the molecular diagnostics of infectious diseases.

The movement state of obstacle including position, velocity, and yaw angle in the real traffic scenarios has a great impact on the path planning and decision-making of autonomous vehicle. Aiming at how to get the obstacle’s movement state in the real traffic scenarios, an approach is proposed to detect and track obstacle based on three-dimensional Light Detection And Ranging (LiDAR). Firstly, the point-cloud data produced by three-dimensional LiDAR after the road segmentation is rasterized, and the reuse of useful non-obstacle cells is carried out on the basis of the rasterized point-cloud data. The proposed eight-neighbor cells clustering algorithm is used to cluster the obstacle. Based on the clustering result, static obstacle detection of multi-frame fusion is worked out by combining real-time kinematic global positioning system data and inertial navigation system data of autonomous vehicle. And we further use the static obstacle detection result to detect moving obstacle located in the travelable area. After that, an improved dynamic tracking point model and Kalman filter are applied to track moving obstacle stably, and we finally get the moving obstacle’s stable movement state. A large amount of experiments on the autonomous vehicle developed by us show that the method has a high degree of reliability.

Protein biomarkers are indicators of many diseases and are commonly used for disease diagnosis and prognosis prediction in the clinic. The urgent need for point-of-care (POC) detection of protein biomarkers has promoted the development of automated and fully sealed immunoassay platforms. In this study, a portable microfluidic system was established for the POC detection of multiple protein biomarkers by combining a protein microarray for a multiplex immunoassay and a microfluidic cassette for reagent storage and liquid manipulation. The entire procedure for the immunoassay was automatically conducted, which included the antibody–antigen reaction, washing and detection. Alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA) and carcinoma antigen 125 (CA125) were simultaneously detected in this system within 40 min with limits of detection of 0.303 ng/mL, 1.870 ng/mL, and 18.617 U/mL, respectively. Five clinical samples were collected and tested, and the results show good correlations compared to those measured by the commercial instrument in the hospital. The immunoassay cassette system can function as a versatile platform for the rapid and sensitive multiplexed detection of biomarkers; therefore, it has great potential for POC diagnostics.