Abstract: An object detection method is provided. In the method, raw point cloud data including a to-be-detected object is obtained, where the raw point cloud data includes annotation information for the to-be-detected object; instance point cloud data corresponding to the to-be-detected object is extracted from the raw point cloud data by using the annotation information; an object position point is determined from the raw point cloud data, and the raw point cloud data and the instance point cloud data to obtain a fused to-be-detected sample is fused based on the object position point and the to-be-detected object is detected by using the raw point cloud data and the to-be-detected sample.

Abstract: A sorting apparatus is provided. The sorting apparatus includes: at least one set of sorting units; where a sorting unit of the at least one set of sorting units includes: a timing belt, where two ends of the timing belt are fixed; a trolley track arranged in parallel with the timing belt; and a distribution trolley, where the distribution trolley comprises a driving assembly and a guide assembly, the driving assembly comprises a driving motor and a timing belt pulley, the driving motor drives the timing belt pulley to cooperate with the timing belt and move along the timing belt; and the guide assembly of the distribution trolley is configured to cooperate with the trolley track to guide a moving direction of the distribution trolley.

Abstract: A distribution site and a distribution method are provided. The distribution site includes: a building having a storage space configured to store goods and a distribution interface area configured to park an unmanned ground distribution vehicle that is capable of performing goods distribution; and a goods sorting device arranged in the building and configured to load the goods in the storage space onto the unmanned ground distribution vehicle.

Abstract: Provided are a sorting vehicle, a goods sorting system and a goods sorting method. The sorting vehicle includes a vehicle body, a belt-type conveying mechanism, an enclosing mechanism and a processor. The belt-type conveying mechanism includes a roller-type drive mechanism disposed on the vehicle body and an endless conveyor belt disposed on a roller of the roller-type drive mechanism. The moving direction of the endless conveyor belt is perpendicular to the traveling direction of the vehicle body. The enclosing mechanism is partially or entirely disposed on the endless conveyor belt, and is configured to form at least one enclosing region above the vehicle body. The processor is configured to control the roller-type drive mechanism to drive the endless conveyor belt to move a preset distance to complete unloading of the goods within the enclosing region; and control the roller-type drive mechanism to complete resetting of the enclosing region.

Abstract: The present disclosure relates to an automatically adjustable smart shelf, a cargo storing method, and an unmanned logistics system. The automatically adjustable smart shelf includes a shelf outer frame and a control unit. A first movable shelf layer and a second movable shelf layer provided on the shelf outer frame and respectively movable along a first direction and a second direction intersecting with each other. A control unit can control the first movable shelf layer and the second movable shelf layer to move along the first direction and the second direction respectively, so as to form various sizes of storage space for storing various cargos accordingly.

Abstract: Provided are a model generation method and apparatus, an object detection method and apparatus, a device and a storage medium. The method includes acquiring multiple scaling coefficients of a batch normalization layer in an initially-trained intermediate detection model, where the intermediate detection model is obtained by training an original detection model based on multiple training samples, and each training sample includes a sample image and a sample annotation result of a known object in the sample image; screening a to-be-pruned coefficient from the multiple scaling coefficients according to values of the multiple scaling coefficients; and screening a to-be-pruned channel corresponding to the to-be-pruned coefficient from multiple channels of the intermediate detection model and performing channel pruning on the to-be-pruned channel to generate an object detection model.

Abstract: The present disclosure provides a control method of a UAV, a control device of a UAV, and a computer-readable storage medium, and relates to the technical field of UAVs. The control method of a UAV includes: determining a deviation between a vertical mapping point on the ground and a landing point of the UAV, the deviation comprising a deviation in a horizontal axis direction of a camera coordinate system and a deviation in a vertical axis direction of the camera coordinate system; and generating speed control amounts of the UAV in the horizontal axis direction and the vertical axis direction of the camera coordinate system by a controller, using the deviation in the horizontal axis direction and the deviation in the vertical axis direction.

Abstract: A method and apparatus for carrying a shelf. One specific implementation of the method comprises: receiving first task information from a task server (104); determining, according to an identifier of a shelf to be transported, the position of a shelf storage space (201) for storing the shelf as a starting point position; determining, according to an identifier of a first picking station, the position of a first buffer storage space in a buffer storage space set corresponding to the first picking station as an ending point position; generating first transporting information according to the determined starting point position and ending point position; and sending the first transporting information to automated guided vehicles (106,107), so that the automated guided vehicles (106,107) transport the shelf.

Abstract: A distribution site and a distribution method are provided. The distribution site includes: a building having a storage space configured to store goods and a distribution interface area configured to park an unmanned ground distribution vehicle that is capable of performing goods distribution; and a goods sorting device arranged in the building and configured to load the goods in the storage space onto the unmanned ground distribution vehicle.

Abstract: A delivery method, device, system, unmanned vehicle, and computer-readable storage medium, relating to the field of logistics technology. The delivery method includes monitoring a state of the unmanned vehicle; verifying a verification code input by the user and received from the unmanned vehicle in response to the unmanned vehicle being not in an abnormal working state, and transmitting an unpacking instruction to the unmanned vehicle in response to successful verification.

Abstract: Disclosed are an information output method and apparatus. One specific implementation of the method comprises: acquiring order data to be sorted; determining whether items to be sorted matching the order data to be sorted are stored in a shelf set; in response to having determined that items to be sorted matching the order data to be sorted are stored in the shelf set, determining, from the shelf set, a shelf storing an item to be sorted, and adding same to a candidate shelf set; choosing a candidate shelf from the candidate shelf set and adding same to a target shelf set, where the target shelf set stores the items to be sorted matching the order data; and outputting an identifier of a target shelf in the target shelf set.

Abstract: The present disclosure provides a vehicle behavior prediction method and apparatus, an electronic device, and a computer-readable storage medium, which relates to the field of intelligent driving. The method includes: determining a target traffic light corresponding to t a target vehicle a current intersection; determining a neighboring traffic light corresponding to the target traffic light and the indication status of the neighboring traffic light according to the target traffic light, the indication status of the target traffic light and a traffic light status mapping relationship table; acquiring a position of an obstacle vehicle and predicting each possible travel path of the obstacle vehicle, and determining a traffic light corresponding to each possible travel path; and determining a final possible travel path of the obstacle vehicle according to each possible travel path and the indication status of the traffic light corresponding to each possible travel path.

Abstract: The present disclosure discloses a dispatching method and device, and a non-transitory readable storage medium, and relates to the field of computer technology. The method of the present disclosure includes: obtaining path condition information within a warehouse; calculating pickup time of each candidate robot of a plurality of candidate robots according to a location of the candidate robot and the path condition information; dispatching a target robot to perform a pickup task according to the pickup time of each candidate robot.

Abstract: A method for determining passage of an autonomous vehicle includes: acquiring information about an intersection on a driving route of the autonomous vehicle, wherein the information about the intersection comprises lane data; acquiring a historical trajectory of an obstacle in the intersection within a specific time; acquiring, by matching the historical trajectory of the obstacle with center lines of respective lanes in the lane data, a lane with smallest matching error; and determining that the lane with the smallest matching error is in a passable state, wherein the lane with the smallest matching error is a lane where the obstacle is located.

Abstract: An article storage and retrieval apparatus includes a main frame; an accommodating assembly, having an article storage state and a folded state, the accommodating assembly including a closed frame rotatably arranged on the main frame and an openable door, and the closed frame being configured to enclose an article storing space between the accommodating assembly and the main frame in the article storing state; a first locking assembly, configured to realize locking between the closed frame and the main frame and allow opening of the article storage and retrieval apparatus after unlocking; and a second locking assembly, configured to realize locking between the closed frame and the openable door and allow the accommodating assembly to switch from the folded state to the article storage state after unlocking.

Abstract: The present disclosure is directed to a pallet assembly, sorting robot and warehouse logistics system. The pallet assembly includes a flap assembly including a first flap and a second flap. The flap assembly has a first state in which the first flap and the second flap are coplanar so as to form a carrying surface for carrying goods, and a second state in which any one of the first flap and the second flap is turned over so as to make the goods slide off. A sorting robot including the pallet assembly and a warehouse logistics system including the sorting robot are also provided.

Abstract: The present disclosure relates to a blockchain-based host security monitoring method and apparatus, a computer readable medium and an electronic device. The host security monitoring method in the embodiments of the present disclosure comprises: monitoring traffic data of a host in network communication, and determining whether the traffic data is malicious traffic; if the traffic data is malicious traffic, obtaining security state information of the host, and saving the security state information to a security state blockchain; generating an invasion log corresponding to the malicious traffic, and saving the invasion log and the security state information to a log storage blockchain.