108results about How to "Expand the range of perception" patented technology

The invention discloses a method for curve synergy collision avoidance early warning based on dedicated short range communication. The method specifically comprises the steps that real-time position information of an automobile of a user is obtained and whether the automobile enters a curve zone or not is judged with the combination of an electronic map; if the automobile enters a curve, traveling information of the automobile of the user is broadcasted to the surroundings and whether an automobile traveling oppositely exists or not is inquired through a dedicated short range communication device; if the automobile traveling oppositely exists, traveling information and position information of the automobile traveling oppositely are obtained and combined with the traveling information and the position information of the automobile of the user, the smallest lateral distance and time needed when the two automobiles meet are predicated, and the smallest lateral distance and the time needed when the two automobiles meet are compared with a set threshold value so that collision avoidance early warning in levels can be achieved. The method for curve synergy collision avoidance early warning based on dedicated short range communication has the advantages that the smallest lateral distance and the time needed when the two automobiles meet can be predicated, interaction of the automobile conditions and the early warning information is achieved through dedicated short range communication, and therefore the purpose of curve synergy collision avoidance early warning is achieved.

The invention discloses a blindman outdoor support system based on computer vision, which comprises an image acquiring unit, a voice output unit, and an image-voice converting unit respectively connected with the image acquiring unit and the voice output unit, wherein the image-voice converting unit comprises an image-voice converting hardware unit and an image-voice converting software unit; and the image-voice converting software unit comprises camera parameter calculation software, plane detection software, obstacle detection and analysis software and image information-voice conversion software. In the system, a binocular camera is utilized to acquire images, a walkable area as well as obstacles and properties of the obstacles corresponding to the physical space in an image are extracted by a software algorithm in an embedded image processor, information is fed back to the blindman through a voice interface so as to assist the blindman to better understand road scenes and make a more comprehensive judgment, expand the sensing range of the blindman and avoid dangers. The device has a small size and low cost, is convenient to carry and facilitates the outdoor activities of the blindman.

The embodiment of the invention discloses an automatic vehicle driving method and a device based on vehicle-road coordination. The method comprises steps of obtaining current real-time location information and destination information uploaded by a target vehicle; obtaining real-time road condition information by using a preset roadside sensor, generating a real-time driving suggestion for the target vehicle according to the current real-time location information, the destination information and the real-time road condition information; and controlling the automatic driving of the target vehicle according to the driving suggestion received by an on-board device and a high-precision map. The automatic vehicle driving method based on vehicle-road coordination can acquire real-time road condition information in a remote detection area by acquiring the roadside sensor, thereby avoiding installing lots of sensors on an automatic driving vehicle, improving automatic driving perception range of the vehicle based on vehicle-road coordination, and enhancing efficiency and safety of the automatic driving vehicle.

The invention provides an intelligent vehicle target tracking system through integration of road side equipment and a vehicle sensor and a method thereof. Vehicle-mounted equipment comprises a target detection module, a vehicle positioning module, a DSRC communication module and a target tracking module. The road side equipment comprises a DSRC communication module and a vehicle tracking module. The target detection module of the vehicle-mounted equipment detects position information of vehicles, pedestrians and targets around a vehicle owner through a vehicle-mounted sensor. The DSRC communication module transmits position and identity information of the vehicle to the road side equipment and receives the position and identity information of the nearby target vehicles transmitted by the road side equipment. The target tracking module performs data association according to target observation data detected by the target detection module, then corrects the association result according to the information transmitted by the road side equipment and finally performs tracking filtering on the corrected association result. The DSRC communication module of the road side equipment receives the position and identity information from various vehicles and transmits the position and identity information to all the vehicles through correction of the vehicle tracking module. Intelligent vehicle target tracking precision can be enhanced by the intelligent vehicle target tracking system through integration of the road side equipment and the vehicle sensor and the method thereof.

The invention aims at the problem how to reproduce the force touch of image textures and provides a method for reproducing the texture force touch based on a shape-from-shading technology from grey scales of an image. The method is characterized by processing the grey scales of a texture image; extracting the micro three-dimensional contour information of the true textures from the grey scale information of the image to be drawn into a virtual surface; rendering the textures of the virtual surface by a force touch model; and outputting calculated force from a general manual controller to an operator. The invention has the advantages that the miniature contour of an object surface is needless to be measured by a special instrument, a special texture force touch expression device is needless to design, and the like and can be used in the fields for simulating virtual environments, controlling remotely-operated robots, browsing remote museums and online shops, simulating surgeries, and the like.

The invention discloses a convolutional neural network training method and apparatus for human face identification and an application, and belongs to the field of human face identification. The method comprises: constructing a sample training library, wherein the sample training library comprises a plurality of sample classes, and the sample classes comprise human face image samples same in quantity; training a convolutional neural network by using the sample training library; extracting eigenvectors of all the human face image samples in the sample training library by using the trained convolutional neural network; classifying the eigenvectors by using a classifier; calculating classification accuracy of each sample class; and judging whether the convolutional neural network meets a set requirement or not, if yes, ending, otherwise, deleting a certain quantity of correctly classified human face image samples from the sample class with highest classification accuracy, adding the same quantity of human face image samples into the sample class with lowest classification accuracy, and re-training the convolutional neural network. The method avoids identification errors caused by influences of facial makeup and external environment, and avoids over-fitting.

The invention discloses a data fusion method and related equipment. The data fusion method comprises the steps of obtaining vehicle sensing data, wherein the vehicle sensing data is obtained by detecting the road environment within a sensing range by a vehicle sensing device through a vehicle sensor; obtaining roadside sensing data, wherein the roadside sensing data is obtained by detecting the road environment within a sensing range by a roadside sensing device through a roadside sensor; and fusing the vehicle sensing data and the roadside sensing data according to a fusion formula so as to obtain a first fusion result. The scheme of the invention can realize the superposition of the sensing range of the roadside sensing device and the sensing range of the vehicle sensing device, therebyeffectively expanding the sensing range.

The invention discloses a detection-information-fusion-based wireless sensor network deterministic deployment method, which comprises the following steps of: detecting the signal strength x of a target by utilizing sensor devices, wherein the target can be detected by a plurality of sensor devices; computing an estimation value of a target characteristic parameter Theta and a parameter estimation error according to the detected signal strength and characteristic parameter expression of the target by adopting a best linear unbiased estimation method; and computing a maximum spacing distance dmax between adjacent sensor nodes; and computing the number of the sensor nodes required to be deployed and specific arrangement positions according to the dmax. In the method, the number of the sensor nodes for detection can be greatly reduced based on a parameter estimation theory under the condition of given detection probability, detection efficiency can be improved and the cost of implementing effective monitoring by a wireless sensor network can be reduced by utilizing the cooperative information processing capability of the sensor nodes; and the method is applied in the fields of factory automatic equipment state monitoring and fault diagnosis, urban traffic safety monitoring and the like.

The invention relates to an ADAS and V2V combined beyond visual range sensing and publishing method and system. The sensing and publishing method comprises data integration publishing process and integrated data receiving handling process; the data integration publishing process utilizes environment sensing data published and integrated by a V2V module, the integrated data receiving handling process utilizes the integrated data and performs collision warning or risk prompt by using a scene classification handling algorithm. By performing state inverse, data filtering and data integration on the collected data of the traditional ADAS vehicle with the V2V module, the sensing and application range of the equipped V2V module vehicle is improved, the application and promotion progress of the V2V module is greatly promoted from the technical angle. The sensing range of the vehicle and the reliability of the V2V application can be improved, and the dependency on an installation rate by the V2V application is reduced.

The invention discloses a ghost probe accident early warning system based on information interaction. The ghost probe accident early warning system consists of a vehicle front detection module, an arduino development board, a location positioning module, a communication module, a vehicle interior voice alarm module, a vehicle external voice alarm module and a vehicle-mounted monitor; a pedestriandetection module consists of two ultrasonic radars which are respectively installed in the left side and the right side of a vehicle head; the output ends of the two ultrasonic radars and the locationpositioning module are both electrically connected with the signal input end of the arduino development board; the control end of the arduino development board is connected with the input end of thevehicle interior voice alarm module, the input end of the vehicle external voice alarm module and the input end of the vehicle-mounted monitor in a control manner; and the communication module is in two-way connection with the arduino development board. Compared with the prior art, the ghost probe accident early warning system can realize the two-way early warning of pedestrian and vehicles and eliminate the safety hazards of pedestrian blind areas, so that the occurrence of ghost probe accidents is effectively avoided and a safety coefficient is high; and in addition, the perceived range of vehicles is expanded.

The invention discloses an automatic driving road condition navigation method and system based on a 5G network and a medium, and is used for solving a technical problem that a vehicle navigation range is limited due to the fact that an existing automatic driving navigation method cannot monitor a long-distance area. The method comprises the following steps of establishing wireless communication connection between roadside equipment and an automatic driving vehicle; acquiring a plurality of frames of obstacle data corresponding to the current monitored road section, and determining a corresponding obstacle type; determining a position or a moving path of the obstacle, and determining an avoidance strategy of the autonomous vehicle on the current monitored road section; according to the preset navigation route, determining a next monitoring road section of the current monitoring road section where the automatic driving vehicle is located and roadside equipment corresponding to the next monitoring road section so as to obtain the number of vehicles in the next monitoring road section; and calculating vehicle density of the next monitored road section, and determining whether the next monitored road section is congested or not according to the vehicle density so as to determine a scheduling strategy for the next monitored road section.

The invention discloses a body joint distributed information consistency estimation method based on interacting multiple models. The method comprises the following steps: initializing a skeleton joint position; estimating joint motion by use of a local sensor; constructing a motion model and an observation model by virtue of a human body joint, and realizing effective estimation on a joint state; estimating information consistency of a target joint between the sensors, i.e., each sensor transmits a joint information vector and an information matrix which are estimated by the sensor and corresponding information contribution and model probability to an adjacent communication sensor node and receives information of the surrounding sensors; by virtue of utilizing an information consistency algorithm and fusing the estimation results of the surrounding sensors, subsequent iteration is performed for multiple times, and convergence of the algorithm and the estimation result is realized; by virtue of constructing a distributed RGBD sensor network, and utilizing the information consistency algorithm, distributed estimation of human body joints is realized; no data fusion center exists in a network, so that robustness of a system for node information mistake and invalidity is improved, and expansion of the sensor network is relatively easily realized.

The invention provides a V2I-based intelligent networked auxiliary perception streetlamp system for autonomous vehicles. The system comprises a streetlamp set in a road network and a data center, wherein the streetlamp is internally provided with a camera image sensor, a millimeter wave radar sensor and a laser radar sensor; the sensors are all connected with the data center to transmit the respectively acquired data to the data center; the data center carries out data extraction and fusion on the basis of a computer vision technology and a data fusion technology after receiving the data, so as to form real-time data of the road network; and the data center is combined with a 5G network service provider to distribute the real-time data of the corresponding road network to each vehicle-mounted terminal through a 5G network according to the demands of the autonomous vehicles, so as to realize the intelligent networked communication of V2I. The system is capable of detecting various pieces of information in the covered traffic range through the road network, so that the vehicle-mounted terminals can conveniently obtain the instant data information on the paths, thereby infinitely enlarging the perception range and avoiding potential safety hazards and traffic jam.

The invention provides a multi-laser radar fusion method based on vehicle-road cooperation, belonging to the technical field of automobiles. The invention aims to solve the problem that accurate and complete target data cannot be provided for people in the prior art. The multi-laser radar fusion method based on vehicle-road cooperation comprises the following steps: performing point cloud processing on target information detected by laser radars installed at a road end and a vehicle end to obtain point cloud data corresponding to the laser radars; conducting clustering, tracking and space synchronization processing on the point cloud data in sequence to form target lists corresponding to the laser radars; performing time synchronization processing on each target list to obtain each targetlist at the same moment; extracting all the point cloud data corresponding to the same target in each target list at the same moment; and performing clustering and tracking processing on all the extracted point cloud data in sequence, and then performing fusion to form a final target list. According to the invention, sensing information with higher precision and better stability can be obtained.

The invention discloses a non-contact sensing method based on an Internet of Things LoRa signal, and belongs to the field of intelligent sensing. According to the non-contact sensing method, the Internet of Things communication technology LoRa with the advantages of low power consumption and long distance is applied to non-contact long-distance sensing, through multiple antennas configured on a LoRa gateway, a method of utilizing a ratio of received signals on two antennas to eliminate phase shift caused by asynchronous receiving and transmitting and reduce noise interference is proposed, andthus the sensing range is increased. On the basis, a phase change measurement method based on a tangent vector is provided, and the phase change of a signal caused by the movement of a sensed human body can be accurately calculated. According to the non-contact sensing method, widely used Internet of Things communication facilities are fully utilized, no extra detection device needs to be installed, and indoor and outdoor non-contact sensing application can be realized in a convenient manner. The non-contact sensing method has accurate perception capability for large-scale walking and small-scale breathing detection of the human body, is easy to popularize, and is suitable for the fields of healthy smart home, smart buildings and the like.

The invention discloses a sensor network based mobile robot and static sensor task cooperation method, which comprises the steps that a dynamic distributed sensor network is built, and consistency estimation is carried out on acquired human body articulation point distributed information; color image and depth image based multi-feature joint data association is performed after consistency estimation is carried out on the human body articulation point distributed information; distributed DTW human body behavior intension recognition is carried out based on human body behavior characteristics; on the basis of a behavior recognition result, a mobile robot participates in a bidding service by using a contract net method and provides robot services in allusion to specific behaviors of a user; when an estimation error of a static sensor for a certain articulation point of the human body exceeds a set value, a request for a close-range checking task is initiated, an adjacent mobile robot bids for the task and executes the task, and close-range observation for a specific target is realized. The sensor network based mobile robot and static sensor task cooperation method provides close-range observation in allusion to an observation blind area on the one hand, and provides corresponding intelligent services for special behaviors of the human body on the other hand.