BEIJING GEEKPLUS TECHNOLOGY CO., LTD.

The provided is a multi-view-based 3D (three-dimensional) object detection method and system. This method is optimized based on an existing three-dimensional object detection model PETR. Specifically, 3DRoPE is introduced to replace an original 3D position encoding mode, and learnable parameters are set for 3D position information to enhance adaptability. In addition, the pose geometric information of multi-view cameras is fused into the position embedding of each view image, so that the capability of this model for processing the complex spatial relationship is further improved. After being trained using the NuScenes dataset, this model can receive real-time image input from a plurality of cameras and output accurate 3D object detection results, which significantly improves perception accuracy and safety in autonomous driving environments.

A method for controlling motion of a robot includes: obtaining a moving scene image collected by a visual sensor, and detecting a target reference object according to the moving scene image to obtain a detection result of the target reference object, wherein the target reference object comprises reference objects at two sides of a moving track of a target robot; determining an edge line of the moving track according to the detection result of the target reference object; determining positioning information of a target shadow eliminating point according to the edge line; and determining a motion adjustment parameter of the target robot according to the positioning information, and adjusting a motion state of the target robot according to the motion adjustment parameter.

A navigation method based on ground texture images, an electronic device and storage medium. The method includes: performing transform domain based image registration on an acquired image of a current frame and an image of a previous frame, and determining a first pose of the image of the current frame; determining whether the image of the current frame meets a preset condition, and if so, inserting the image of the current frame as the key-frame image into a map, and performing loop closure detection and determining a loop key-frame image; performing transform domain based image registration on the image of the current frame and the loop key-frame image, and determining a second pose of the image of the current frame; and determining an accumulated error according to the first pose and the second pose, and correcting the map according to the accumulated error.

An intelligent forklift and a method for detecting a pose deviation of a container are provided. The intelligent forklift includes a working state monitor, an image sensor, and a processor. The working state monitor is configured to monitor a working state of the intelligent forklift while the intelligent forklift is carrying and moving a stock container and send a trigger to the image sensor based on the working state. The image sensor is configured to capture an RGBD image frame containing the stock container in response to the trigger and to transmit the RGBD image frame to the processor. The processor is configured to receive the RGBD image frame, detect a pose deviation degree of the stock container and provide an alarm and/or a prompt based on a detection result of the pose deviation degree for adjusting the pose of the stock container.

An intelligent forklift and a method for detecting a pose deviation of a container are provided. The intelligent forklift includes a working state monitor, an image sensor, and a processor. The working state monitor is configured to monitor a working state of the intelligent forklift while the intelligent forklift is carrying and moving a stock container and send a trigger to the image sensor based on the working state. The image sensor is configured to capture an RGBD image frame containing the stock container in response to the trigger and to transmit the RGBD image frame to the processor. The processor is configured to receive the RGBD image frame, detect a pose deviation degree of the stock container and provide an alarm and/or a prompt based on a detection result of the pose deviation degree for adjusting the pose of the stock container.