Abstract: The present invention relates to a vision-language large model-based multi-robot collaborative navigation method and system. The method includes steps of: according to a task instruction input by a user, combined with a semantic map constructed by each robot, allocating subtask instructions to robots after parsing by a large language model; acquiring, by each robot, an environmental image in real time, according to the allocated subtask instruction, parsing and predicting a next action using a vision-language navigation large model, and executing a corresponding action and updating a state by the robot; and monitoring a robot state and checking task progress in real time through a human-machine interface and adjusting a task or handling an exception during execution when needed by an operator, and dynamically adjusting, by the robot, an execution strategy according to feedback information to optimize navigation for task completion.

Abstract: The present invention relates to a robot action generation method and system combining general and specialized models, where the method includes: constructing the general model and the specialized model, pre-training the general model, performing parameter fine-tuning on a pre-trained general model, and training the specialized model based on a fine-tuned general model; acquiring a task instruction and real-time visual information, inputting the task instruction and real-time visual information into the fine-tuned general model, and outputting an action sequence and a task latent feature; and acquiring real-time point cloud perception data, and inputting the real-time point cloud perception data, together with the action sequence and the task latent feature, into a trained specialized model, and outputting continuous robot actions. Compared with the prior art, the present invention improves the speed of robot action generation and enhances the generalization of robot action generation.

Abstract: A method for training a robot action generation model and a method for generating robot actions. The training method includes: constructing an automotive wire harness operation dataset and extracting human poses; constructing an instruction set based on the dataset and the human poses; pre-training a motion tokenizer; obtaining a text-motion vocabulary; pre-training a language model based on the dataset, the instruction set, and the text-motion vocabulary; and constructing an operation dataset and fine-tuning the pre-trained language model based on the operation dataset to complete training. The method for generating robot actions utilizes the robot action generation model trained by the training method to output a humanoid action sequence and redirects the humanoid action sequence to a robot to generate robot actions. The present disclosure enhances generalizability of the generation model, makes the generation of robot actions more aligned with automotive wire harness operations.

Abstract: Provided are a method and system for testing a short-circuit withstand capability of a transformer, a terminal, a medium, and a program. The method includes: acquiring a transformer design parameter and an operating condition of a power system where a transformer is located; acquiring a maximum short-circuit current passing through a transformer winding based on the transformer design parameter and the operating condition of the power system; inputting the maximum short-circuit current passing through the transformer winding into a first preset calculation model, and acquiring a first stress value for each disc of the transformer winding; acquiring a second stress value for each disc of the transformer winding by using a first predetermined method; acquiring an allowable current of the transformer winding by using a second predetermined method; acquiring a safety factor of the transformer, and determining a short-circuit withstand capability of the transformer.

Abstract: The present disclosure provides a multi-view collaborative tracking method and apparatus for a fast moving target under a low luminance condition, including a multi-view target tracking algorithm based on radar-vision fusion data and an unmanned aerial vehicle (UAV) cluster collaborative tracking technology. This method takes the spatiotemporal normalized multi-source data as the learning objective of multi-agent reinforcement learning, to drive the cluster to collect data from multiple angles and directions. In this way, highly-efficient multi-angle collaborative observation can be performed on a high-speed moving object in an environment of weak texture and low luminance.

Abstract: Provided is a method for token-based representation and learning of robotic perception data based on a graph neural network, comprising: obtaining a plurality of types of perception data of a robot; performing token-based representation according to types of the plurality of types of perception data; constructing an initial feature graph based on the plurality of types of perception data after the token-based representation; learning a compact representation of the initial feature graph based on an autoencoder and reconstructing a graph structure; after the autoencoder completes learning of the graph structure, fixing the graph structure; and converting the plurality of types of perception data into node feature vectors, constructing a feature graph based on the graph structure, and performing numerical encoding on each of the node feature vectors by utilizing the graph neural network to obtain a representation of high-dimensional feature vectors of the plurality of types of perception data.

Abstract: A method for snake robot navigation, the method including: providing a snake robot comprising a plurality of modules and a plurality of tactile sensors disposed thereon; planning a path over a terrain between an initial position and a target location; detecting a tactile datum from the plurality of tactile sensors; selecting, based on the path, one of a plurality of gaits the snake robot may perform by relative movement of the plurality of modules; dynamically adjusting the selected gait based on the tactile datum; and commanding the plurality of modules to perform the adjusted gait.

Abstract: A method and device for storing information in a molecule. The method comprises: obtaining initial information to be stored, and representing the initial information using a first address code and a first content code; recoding each first address code respectively, to represent one respective first address code using first recoding information with a first preset bit number and a first preset base; determining respective molecular modules according to the first content code and the first recoding information, and combining the determined molecular modules in order, such that a composition corresponds to the initial information.

Abstract: A battery housing includes a bottom case and a top cover. The bottom case includes a main case and a reinforcing rib. The main case is provided with an accommodating cavity and an opening, the accommodating cavity being located at an inner side of the main case, and the opening being located on the outer surface of the main case and communicated with the accommodating cavity. The reinforcing rib is fixedly connected to a side wall surface of the accommodating cavity, is spaced apart from both the opening and the bottom wall surface of the accommodating cavity, and extends along the circumferential direction of the main case. The first sub-reinforcing portion of the reinforcing rib includes a step surface facing the opening, the second sub-reinforcing portion is located on a side of the first sub-reinforcing portion facing away from the step surface and fixedly connected to the first sub-reinforcing portion.

Abstract: A method and device for storing information in a molecule. The method comprises: obtaining initial information to be stored, and representing the initial information using a first address code and a first content code (S100); recoding each first address code respectively, to represent one respective first address code using first recoding information with a first preset bit number and a first preset base (S210); determining respective molecular modules according to the first content code and the first recoding information (S310), and combining the determined molecular modules in order, such that a composition corresponds to the initial information (S400).

Abstract: The present application discloses an anthropogenic heat flux estimation method and system based on flux observation data. The method includes: performing grid division on the selected region, and acquiring building vector data, land utilization data and instrument observation data in the selected region; calculating and correcting sensible heat fluxes and latent heat fluxes of different underlying surface types in a vorticity correlator flux source region; calculating temporal-spatial distribution of the sensible heat flux and the latent heat flux of each grid in a large-aperture scintillometer flux source region, and calculating temporal-spatial distribution of an anthropogenic heat flux of each grid in the flux source region in combination with a surface energy balance equation; and constructing an anthropogenic heat flux estimation model based on a convolutional neural network, and accurately estimating temporal-spatial distribution of the anthropogenic heat flux of each grid in the selected region.

Abstract: The present disclosure provides a multi-view collaborative tracking method and apparatus for a fast moving target under a low luminance condition, including a multi-view target tracking algorithm based on radar-vision fusion data and an unmanned aerial vehicle (UAV) cluster collaborative tracking technology. This method takes the spatiotemporal normalized multi-source data as the learning objective of multi-agent reinforcement learning, to drive the cluster to collect data from multiple angles and directions. In this way, highly-efficient multi-angle collaborative observation can be performed on a high-speed moving object in an environment of weak texture and low luminance.