The invention discloses a positioning method and device in an indoor dynamic scene and relates to the field of indoor positioning. The method comprises a step of acquiring a current frame in real timein the indoor dynamic scene, a step of determining matched image features between the current frame and a previous frame and estimating a transformation matrix between the current frame and the previous frame according to the matched image features, a step of calculating the matching percentage of the current frame and the previous frame according to the transformation matrix and updating a global pose by adopting an axle odometer when the percentage is smaller than a specified threshold value, and a step of detecting whether a key point is passed or not in a movement process and refining theglobal pose if so. The device comprises an obtaining module, a matching module, an updating module and a refining module. According to the invention, the indoor positioning is carried out in the presence of obstacles and dynamic events, the positioning information is optimized through the key point of a known space, and a more optimized result is obtained in a specific scene in combination with the axle odometer.

Positioning method and device in indoor dynamic scene

The application, which belongs to the technical field of robots, particularly relates to a robot pose estimation method and device, a computer readable storage medium and a robot. The method comprisesthe following steps: acquiring inertial measurement data of a robot, and performing pose estimation according to the inertial measurement data to obtain a reference estimated pose of the robot; aquiring absolute measurement data of the robot respectively, carrying out pose estimation according to the absolute measurement data to obtain absolute estimation poses corresponding to the absolute measurement sensors; acquiring relative measurement data of the robot respectively, and performing pose estimation according to all relative measurement data to obtain a relative estimation pose corresponding to each relative measurement sensor; and performing fusion processing on the reference estimated pose according to the absolute estimated pose and the relative estimated pose to obtain a fused pose of the robot. According to the application, the precision of robot pose estimation is greatly improved.

Robot pose estimation method and device, readable storage medium and robot

The invention discloses a flexible networking communication platform for heterogeneous equipment. The platform comprises two functional modules, namely an equipment integration-oriented flexible networking design module and a workshop operation process-oriented inter-equipment real-time communication service module. The equipment integration-oriented flexible networking design module operates in the foreground in the form of a visual human-computer interface, and the inter-equipment real-time communication service module for the workshop operation process operates in the background in the formof sub-threads to concurrently process data transmission among terminals in real time. The invention provides flexible networking communication platform software for heterogeneous equipment, so thatdependence on equipment manufacturers and interface types can be reduced when enterprises build production workshops, agile design and reconstruction services can be provided for equipment networkingschemes of different industrial sites, and in addition, stable and efficient real-time communication services among equipment in the workshop operation process can be guaranteed by software.

Flexible networking communication platform for heterogeneous equipment

The safety of pedestrian crossing has always been an important issue in the field of traffic safety. However, at present, vehicle road coordination early warning still needs to be improved. Therefore, a millimeter-wave radar vehicle-road collaborative pedestrian warning device using edge computing modules is proposed, which can detect pedestrians in real time, efficiently and accurately, then alarms drivers. The device includes five modules: radar data acquisition module, edge computing module, data transfer module, cloud server, and data display module. The system architecture is different from the traditional architecture. It combines artificial intelligence technology and the Internet of Things, and delegates a large amount of computing work to the edge computing module to obtain better system performance. Moreover, a RSU (road side unit) is designed to store, process, and upload the data processed by the edge computing module near the intersection in real time, instead of directly transmitting the data to the cloud server. In the actual deployment, this device is proved that it is able to have a good and stable detection in dark night environment and complex weather. 

A Pedestrian Sensing Warning System with Edge Compute Module on Millimeter Wave Radar

The invention provides a positioning and navigation method, a positioning and navigation device and a storage medium, wherein the method comprises the following steps: performing primary positioning for a target carrier in a monitoring range, calculating motion information of the target carrier under a geographic coordinate system, and converting the motion information into the motion informationunder a world coordinate system, obtaining the current position coordinate and speed of the target carrier according to the converted motion information, obtaining motion trail according to the current position coordinate and the speed, acquiring beacon position information of a beacon preset in the monitoring range, obtaining relative position data between the target carrier and the beacon according to the beacon position information and the current position coordinate of the target carrier, and regulating the motion trail of the target carrier according to the relative position data. According to the method and the device provided by the invention, the current position coordinate, the speed and the motion trail of the target carrier in the geographic coordinate system are obtained through calculation of the motion information of the target carrier, the motion trail of the target carrier is regulated according to the relative position of the target carrier and the beacon, and thus, positioning and navigation of the target carrier can be more accurate.

Positioning and navigation method, positioning and navigation device and storage medium

The invention requests the protection of an inertial navigation high-precision positioning method based on binocular, acceleration and gyroscope. The three-dimensional coordinates of the characteristic points of an inertial navigation reference position are determined based on binocular positioning and high-precision GPS information, and the three-dimensional coordinates of inertial navigation areupdated in real time by using the coordinate values of the characteristic points as reference points to eliminate the cumulative error caused by the integration of inertial navigation, and assist inhigh precision positioning of vehicles in an indoor parking lot based on inertial navigation and binocular positioning. A structure block diagram of the invention includes the following parts: I, thecomposition of a high precision inertial navigation system and the structure of a reference coordinate system; II, an acquisition method of the initial coordinates of inertial navigation; III, a selection method of binocular high precision positioning reference points; IV, the selection and design of a high precision inertial navigation coordinate system; V, a conversion method of a high precisioninertial navigation coordinate system; and VI, an elimination method of the cumulative error of inertial navigation.

Inertial navigation high-precision positioning method based on binocular, acceleration and gyroscope

Accurate vehicles counting for all-weather in cities are an important part of traffic management in the application of Intelligent Transportation Systems (ITS). Vehicle counting is currently collected with computer vision and sensor network methods. However, these methods require expensive hardware to achieve real-time and anti-interference capability, and do not provide lane-level vehicle information for ITS traffic management. This paper presents a lane-level vehicle counting system that is based on V2X communications and centimeter-level positioning technologies. This system can be used to traffic survey of ITS at a range of urban intersections. For realizing lane-level counting, a lane determination method is designed with on-board units (OBUs) in this paper. The lane is identified by matching the vehicle positioning information with road information from the roadside unit (RSU). The RSU collects the vehicle counting information from OBUs in different instances. The counting information includes the vehicle location data, the vehicle status data, and the vehicle number of each lane in the range of intersections. Verification and analysis were performed by a hardware-in-the-loop simulation platform. The results showed an average vehicle counting accuracy rate (99.60%). The system enabled the collection of real-time statistics with low-power consumption and low latency, providing accurate data to ITS.

Lane-Level Vehicle Counting Based on V2X and Centimeter-level Positioning at Urban Intersections

The invention discloses a V2X vehicle-mounted terminal message self-adaptive scheduling management system and method, and the system employs a middleware-based software architecture, and mainly comprises a thread pool management module, a message self-adaptive forwarding management module, and a heterogeneous communication network management module. According to the method, heterogeneous multi-source information such as coming vehicle bodies, other vehicles, roadside equipment and an Internet of Vehicles cloud platform is managed by a vehicle-mounted terminal in a unified manner. According to the method, the message is dispatched based on the priority strategy to ensure that the information with higher emergency degree is preferentially sent, so that the real-time performance of the security message is improved; the message parallelization processing is carried out based on the thread pool, and in addition, for the heterogeneous communication network, the heterogeneous network communication is managed by adopting a port-based design method, so that a uniform communication calling interface is provided for an upper-layer application, the heterogeneity is shielded, and the application development efficiency and the function expandability are improved.

V2X vehicle-mounted terminal message self-adaptive scheduling management system and V2X vehicle-mounted terminal message self-adaptive scheduling management method

Jiang Jianchun

Papers

Inertial navigation high-precision positioning method based on binocular, acceleration and gyroscope

Lane-Level Vehicle Counting Based on V2X and Centimeter-level Positioning at Urban Intersections

V2X vehicle-mounted terminal message self-adaptive scheduling management system and V2X vehicle-mounted terminal message self-adaptive scheduling management method