10results about How to "Improve task performance" patented technology

The invention relates to a remote sensing satellite task planning application performance evaluation method. The method comprises the following steps: establishing a hierarchical evaluation index system; definition and quantification of task completion capability indexes are completed; defining and quantifying resource utilization efficiency indexes; defining and quantifying a resource collaboration efficiency index; performing quantitative evaluation on the relative importance degree of each index element in the same level in the hierarchical structure and constructing a judgment matrix; checking the consistency of the judgment matrix; according to the judgment matrix, the proportion of elements in the hierarchical structure in the decision is calculated, and parameter weights including hierarchical single sorting and hierarchical total sorting are determined; and calculating a comprehensive efficiency evaluation value according to each index numerical value and the parameter weight. According to the method, quantitative evaluation of the application efficiency of satellite task planning can be realized, and closed-loop optimization of a task planning function in a remote sensing satellite management and control system is supported.

The invention belongs to the technical field of aircraft trajectory optimization, and discloses an aircraft reentry trajectory intelligent generation method based on an RBF neural network. According to the method, firstly, an aircraft reentry model is constructed, a trajectory optimization problem needing to be solved is constructed based on the aircraft reentry model, an optimal sample trajectory set is generated by using a pseudo-spectral method according to an initial discrete height speed, and then a sample trajectory is generated at each waypoint according to a new discrete state. The sample set is divided according to the waypoints, the RBF neural network is trained, the input of the RBF neural network is a discrete height speed state, and the output of the RBF neural network is a state-control sequence from the current state to a target point; and finally, the RBF neural network is used for online generation of an optimal track. According to simulation verification, the method is high in confidence coefficient and small in error, and the generation efficiency of the optimal track can be greatly improved.

The invention discloses an equipment cluster maintenance decision-making method based on multi-agent deep reinforcement learning, and aims to solve the problem of collaborative maintenance of equipment clusters deployed in a multi-site distributed manner under the condition of dynamic change of task requirements. According to the method, on the basis of equipment health status, performance level and task requirements, an equipment cluster maintenance decision process is modeled as a Markov decision process, a centralized training-distributed execution multi-agent deep reinforcement learning framework is adopted, a local strategy network is constructed for each geographical deployment site, and the maintenance decision process is optimized. And meanwhile, a centralized state value evaluation network is utilized to evaluate the overall operation state of the equipment cluster, so that collaborative optimization of equipment cluster maintenance decisions deployed across sites is realized. By training a multi-agent strategy in a simulation environment and performing online execution in actual operation, each agent can autonomously generate a maintenance strategy based on a local equipment state, and comprehensive optimal control of overall task capability and maintenance cost is realized on a system level. According to the method provided by the invention, the self-adaptive optimization of the equipment cluster maintenance strategy can be realized under the complex conditions of a large number of equipment, high state dimension and dynamic change of task requirements, so that the overall task performance of the equipment cluster is improved, and the maintenance cost is reduced.

A multi-mode autonomous underwater robot, which is a whole body streamlined structure of a rotary body, is composed of a bow load cabin section, a buoyancy adjusting cabin section, a battery cabin section, a navigation cabin section, a control cabin section, a steering cabin section, a propelling cabin section and an antenna in series by sequentially arranging, and is sealed and connected between adjacent cabin sections by a hoop. In the aspects of hardware and structure, the multi-mode autonomous underwater robot has the characteristics of modularity and easy expansion. When the function needs to be expanded, the load sensor in the bow load cabin section can be replaced as needed, all cabin sections adopt the same electrical interface scheme, and the function expansion can also be realized by additionally installing an additional sensor load cabin section. The multi-mode autonomous underwater robot realizes the function compatibility of an AUV, a floating buoy, an ARGO buoy and an underwater glider on the same underwater robot, can realize five operation modes of a water surface navigation mode, an underwater navigation mode, a floating buoy mode, an ARGO buoy mode and an underwater glider mode, the operation modes can be freely switched, and the application range and task capability of the underwater robot are greatly expanded.

The application provides a task execution method and device based on agent and MCP collaborative optimization, and relates to the technical field of intelligent scheduling, wherein the task execution method based on agent and MCP collaborative optimization comprises the following steps: acquiring a predefined graph structure, wherein the graph structure comprises one or more candidate super edges of each type of task, and the nodes in the candidate super edges are model context protocol (MCP) tools or agents for task execution; receiving a task instruction input by a user, and acquiring a context feature vector according to the graph structure and the task instruction; determining the matching scores of each candidate super edge in the graph structure according to the context feature vector; comparing the matching scores of the candidate super edges, and determining a target super edge from the candidate super edges; calling the MCP tools and / or agents corresponding to the target super edge to execute the task instruction, obtaining a task execution result, and realizing intelligent collaboration between multiple agents and MCP tools, thereby efficiently executing the task.

This invention belongs to the field of radar information processing technology and discloses a pre-training method for a multimodal model of radar target detection, tracking, and recognition. The invention includes collecting model pre-training data; converting the collected radar information processing data from various models into a unified data format through data preprocessing to serve as input data for the pre-trained model; achieving simultaneous input of multimodal data, including images, tracks, and category text, through a unified pre-training framework, and completing joint representation learning of the input data by the pre-training framework; and designing a multi-task joint learning optimization algorithm to achieve parallel training of radar target detection, tracking, and recognition tasks. This invention achieves integrated processing of radar target detection, tracking, and recognition tasks by establishing a multi-task learning mechanism under a unified model training framework, effectively improving the overall performance of radar information processing. Simultaneously, by enhancing versatility and generalization, the model is applicable to intelligent processing tasks of various radar models.

The invention relates to the technical field of bionic robots, in particular to a variable-structure snakelike modular triphibian cross-medium bionic specialized robot. The variable-structure snakelike modular triphibian cross-medium bionic specialized robot is composed of eleven functional modules, comprises a head master control module, a joint connection module, a ducted fan module and a tail module, has water, land and air triphibian movement capacity, is driven in a redundant mode through hybrid power, integrates bionic winding, wheel type rolling and a lift force ducted system, and is high in stability and high in reliability. On the basis of CPG distributed control, environment adaptation and path planning are achieved in combination with a visual sensing system, cross-medium switching can be autonomously completed, robustness is improved through modular design, and the robot is suitable for complex environment operation and has high propulsion efficiency and task completion capacity.

This invention provides a lightweight adaptation method and system for large models. By extracting the semantic core from educational data, domain topic vector clusters are obtained, and pseudo-instruction-response pairs are generated to expand the training set based on these clusters. In terms of model structure, a low-rank adapter module is inserted into the transformer layer of the pre-trained large model, and the semantic direction of the topic vector clusters is extracted through singular value decomposition. The adapter parameters are initialized, and a weighted strategy is adopted during training: the prediction entropy and cross-entropy loss of each sample are calculated, and sample weights and gradient adjustment coefficients are generated based on these. The gradient is scaled and updated, and orthogonality constraints are applied, with the strength of the constraints adjusted by the sample weights.

The application relates to a Hilbert fractal curve-based image convolution method and system, comprising the following steps: 1) performing a padding operation on an input image according to a convolution output size mode, for example, the input size is equal to the output mode, and a padding operation needs to be performed on the input edge; 2) block index and rearrangement: performing image block reading and rearrangement operations on the padded input image according to the index sequence number of the Hilbert fractal; 3) cross-correlation operation: performing convolution cross-correlation operation on the rearranged image block with a step equal to the convolution kernel size; 4) local feature aggregation: performing convolution cross-correlation operation on the output of step 3 with a step of 1 by using a deep separable one-dimensional convolution; 5) restoring the two-dimensional structure of the input: performing reverse rearrangement operation on the output of step 4 according to the index of the Hilbert fractal, and outputting. Compared with the existing image convolution operation method, the application is simple, has strong universality and has certain theoretical advancement.

The application belongs to the technical field of aircraft trajectory optimization, and discloses an intelligent generation method of reentry trajectory of an aircraft based on a RBF neural network. Firstly, a reentry model of the aircraft is constructed, and a trajectory optimization problem to be solved is constructed based on the reentry model; then, an optimal sample trajectory set is generated by using a pseudospectral method according to initial discrete altitude and speed; after that, a sample trajectory is generated according to a new discrete state at each waypoint. Then, the sample set is divided according to the waypoints, and the RBF neural network is trained respectively; the input of the RBF neural network is the discrete altitude and speed state, and the output is a state-control sequence from the current state to the target point. Finally, the RBF neural network is used to generate an optimal trajectory online. The simulation verifies that the confidence of the method is high and the error is small, and the generation efficiency of the optimal trajectory can be greatly improved.