59 results about "Mobile agent" patented technology

A method of generating at least one trajectory in scenario comprising an agent navigating a mapped area, the method comprising: receiving an observed state of the agent and map data of the mapped area; generating a set of multiple trajectory basis elements from the observed state of the agent based on the map data; processing one or more scenario inputs in a neural network to generate a set of weights, each weight corresponding to one of the trajectory basis elements; and generating a trajectory for the agent by weighting each trajectory basis element by its corresponding weights and combining the weighted trajectory basis elements.

The invention discloses a mobile terminal agent information processing system based on a prior rule and a multi-modal large model, and relates to the technical field of artificial intelligence, and the system comprises a graph standard determination positioning and recognition module which accurately positions and recognizes icons in an interface through different levels of recognition methods, and deduces the function of the icons; the execution efficiency improving module is used for improving the execution speed through task disassembly and optimization path design; the operation optimization module of the specific application is used for recording the position information, the size and the corresponding icon function of each interface element, constructing a navigation path library of the specific application and optimizing the execution path of the periodic task; and the navigation and task data collecting and updating module is responsible for collecting and updating dynamic data of navigation paths and task execution and continuously performing self-learning and optimization. Therefore, by the adoption of the method, accurate icon positioning and function recognition can be achieved, the execution efficiency is improved, computing resource consumption is reduced, and meanwhile interface changes can be coped with and continuously optimized.

The invention provides a method and device for generating training data of a mobile agent, equipment and a medium, and relates to the technical field of artificial intelligence and the like. According to the specific implementation scheme, the method comprises the steps that multiple pieces of triple data representing interaction behaviors in an application are collected; the triple data comprises a first user interface state, an action and a second user interface state; based on the multiple pieces of triple data, constructing a state transition diagram; obtaining an interaction track based on the state transition diagram; and based on the interaction track, generating training data of the mobile intelligent agent. According to the technology disclosed by the invention, the generation of the training data of the mobile agent can be automatically realized, and the generation accuracy and the generation efficiency of the training data of the mobile agent can be effectively improved.

The invention provides a ground-air obstacle avoidance method guided by a time-varying vector field, and the method comprises the steps: carrying out the clustering of obstacles based on the difference between the motion feature information of all obstacles at any moment, and determining the types of the obstacles based on a clustering result; based on the obstacle type of the obstacle and the movement feature information of the obstacle, determining the threat degree of the obstacle to the position of the mobile intelligent agent at any moment; based on the threat degree of each obstacle, a global time-varying vector field is determined, a navigation vector field of the mobile agent is determined based on the global time-varying vector field, and the global time-varying vector field represents the comprehensive risk of the position of the mobile agent at any moment; and carrying out obstacle avoidance path planning based on the navigation vector field of the mobile intelligent agent. According to the method provided by the invention, the motion features of different types of obstacles are mapped to a unified feature vector field space in a dimensionality reduction manner, the obstacle avoidance problem in a complex dynamic environment can be efficiently solved, and the safe operation of the mobile intelligent agent is ensured.

The invention relates to the technical field of communication and perception fusion, and discloses a joint optimization method and system for moving target tracking and multi-user communication, which are applied to an ISAC system for sensing a moving target and multi-user communication by a mobile agent, and are characterized in that a target state vector is established based on a preset observation model, and the future state of the target is predicted by using extended Kalman filtering; based on a prediction result, using a Cramer-Rao lower bound to quantify a perception error, using a user signal-to-noise ratio as a communication constraint to construct a joint optimization target which minimizes the sum of perception performance lower bounds and satisfies communication QoS, transmitting power and intelligent agent speed constraints, and decoupling the joint optimization target as a beam forming and trajectory optimization sub-problem to be solved through an alternating iteration algorithm; transmitting a signal based on an optimization result, receiving, observing and updating a target state to form closed-loop feedback; and outputting the optimization parameter as the initial input of the next time slot for loop execution, thereby improving the sensing precision, the communication quality and the resource utilization efficiency of the system in a dynamic scene.

A method, apparatus and system for efficient navigation in a navigation space includes determining semantic features and respective 3D positional information of the semantic features for scenes of captured image content and depth-related content in the navigation space, combining information of the determined semantic features of the scene with respective 3D positional information using neural networks to determine an intermediate representation of the scene which provides information regarding positions of the semantic features in the scene and spatial relationships among the semantic features, and using the information regarding the positions of the semantic features and the spatial relationships among the semantic features in a machine learning process to provide at least one of a navigation path in the navigation space, a model of the navigation space, and an explanation of a navigation action by the single, mobile agent in the navigation space.

The invention proposes a computer-implemented method for operating a plurality of mobile agents (26) in an environment, comprising: a) generating mission information that is indicative of any of a number of mobile agents (26), a pickup location, a drop off location, and at least one preset-time, wherein each mobile agent (26) is associated with a depot, wherein the preset-time is indicative of the maximum time period in which the number of mobile agents (26) has to complete a movement phase; b) at least one movement phase including, for each of the number of mobile agents (26), generating a control command (24) based on the mission information (14), wherein the control command (24) causes the respective mobile agent (26) to move from a starting location (16) to a destination in accordance with the mission information (14); wherein each control command (24) is generated such that the respective mobile agent (26) arrives at its destination within the preset-time.

The present disclosure relates to a mobile agent in an RFID system with reduced power consumption and increased localization capabilities. In one aspect, a relay circuit attached to a mobility mechanism such as a drone or robot is in communication with an RFID reader and a backend host computer. The mobile agent moves around a warehouse, for instance, in which landmark RFID tags are arranged. The mobile agent can transmit with an adjustable power to identify a single landmark tag and associate it with object tags of one or more objects within the range of the mobile agent. The host computer can communicate with the mobile agent to instruct it to adjust its power when multiple landmark tags are detected.

The invention discloses a multi-agent priority obstacle avoidance path planning method for a dynamic obstacle, and relates to the technical field of mobile agent path planning, the method comprises the following steps: a global planning stage: optimizing an A * algorithm by introducing a heuristic scaling factor, an obstacle and direction penalty term and a neighborhood adaptive extension mechanism, a path is smoothed by combining broken line reconstruction and a Bezier curve; in the local planning stage, the weight of a DWA evaluation function is dynamically adjusted through fuzzy control, and the real-time obstacle avoidance robustness is improved. Meanwhile, a priority obstacle avoidance strategy and a dynamic look-ahead mechanism are designed, multi-agent conflicts are coordinated, and collective stagnation is avoided. Simulation results show that the method shows better path length, search efficiency and obstacle avoidance performance in static, unknown static and dynamic obstacle environments, effectively improves the multi-agent coordination stability, and is suitable for actual scenes such as warehouse logistics and the like.

A method of predicting trajectories for agents of a scenario, the method comprising, for each agent generating an agent feature vector based on one or more observed past states of the agent, computing a set of pairwise feature vectors, each computed as a combination of the agent feature vector for that agent with a respective agent feature vector generated for each other agent of the scenario, processing the pairwise feature vectors as independent inputs to one or more interaction layers of a trajectory prediction neural network to generate a pairwise output for each pairwise feature vector, aggregating the pairwise outputs over the other agents of the scenario to generate an interaction-based feature representation for each agent, processing the interaction-based feature representation in one or more prediction layers of the trajectory prediction neural network, and generating, based on the output of the one or more prediction layers, at least one predicted trajectory for each agent.

The invention discloses a decomposition type value evaluation multi-agent collaborative navigation method based on an attention mechanism. The method comprises the following steps: modeling a collaborative navigation task of multiple agents into a decentralized partially observable Markov decision process; designing local attention modules of a precoding module and an attention coding stage to extract and code environment feature information of multiple agents, and sharing the local attention modules by a value network and a strategy network; designing a multi-agent strategy-value method under decomposition type value evaluation; constructing a static target navigation environment, and guiding the mobile intelligent agent to actively search for an optimal position; constructing a cooperative hunting environment; a multi-agent cooperative navigation model is constructed, an optimal model is obtained through training and testing, and non-explicit and complete perception navigation of multi-agent cooperation with the reliability distribution problem is achieved. The collaborative navigation method provided by the invention has the advantages of faster convergence, higher success rate, shorter time and less collision, and shows good generalization ability and robustness.

The invention belongs to the technical field of mobile communication, and discloses an air federation learning system and learning method integrating communication, sensing and calculation, and the learning system comprises a model learning module, an equipment sensing module, a mobile agent and a core server. The learning method comprises the steps that firstly, the mobile agents carry out local model training according to the representative weights of the regions and then carry out local model uploading, and the core server senses the position and state information of the mobile agents, updates the representative weights of the regions and sends corresponding power compensation coefficients to all the mobile agents; and the core server carries out global model aggregation and broadcasts the global model to carry out next round of federal learning. In a non-independent identically distributed scene, position information and channel states of all mobile agents are considered, and a global model is updated according to local gradient deviation so as to reduce influence caused by local sample deviation.

The application discloses an identity management method and system, which is suitable for mobile agents or other devices. The method comprises the following steps: generating a plurality of public-private key pairs for each mobile agent before the mobile agent performs a target task, wherein each mobile agent corresponds to an actual identity identifier, and each public-private key pair corresponds to a public-private key pair identifier; determining a plurality of temporary identity identifiers corresponding to the mobile agent according to the actual identity identifier and the public-private key pair identifier; generating a plurality of digital certificates corresponding to the mobile agent according to the temporary identity identifier and the public-private key pair; signing the digital certificate by using the private key of the mobile agent device owner, and storing the signed digital certificate, the temporary private key in the public-private key pair and the public key of the mobile agent device owner into the mobile agent. The application solves the technical problem that the mobile agent cannot indicate that it belongs to the same owner in a privacy protection manner in the related art.

The application discloses a kind of dynamic threshold point cloud matching mobile intelligent agent relative pose cooperative acquisition method, utilize shared laser point cloud data and autonomous positioning result, using improved iterative nearest neighbor point (ICP) registration algorithm, the relative pose between mobile intelligent agent is calculated, and the reliability of calculation result is evaluated, further improve the observation reliability and application value of relative pose.The method of the present application only needs to reuse lidar data on the basis of autonomous positioning, without additional enhancement mark, the observation calculation of relative pose can be completed, the original system is changed little, does not affect the original layout of system, does not need to rely on other feature enhancement hardware, no additional influence, low cost;It has better accuracy and stronger robustness, and the effectiveness evaluation of point cloud registration result makes the application more suitable for the application scene of relative pose, and enhances the reliability of result utilization.

Disclosed is a method for transmitting and receiving autonomous driving-related data, which includes checking, at a second mobile agent, a web server address of a first mobile agent, performing a first identity check procedure between the second mobile agent and the first mobile agent, and forming a social between the second mobile agent and the first mobile agent.

A data processing apparatus (100) and method for efficiently and accurately determining a relative pose (160) of a mobile agent such as a vehicle are disclosed. The apparatus (100) comprises processing circuitry (110) configured to determine a feature vector and a score for each point in a source point cloud (140) and a feature vector for each point in a target point cloud (150), and to transform a plurality of key points of the source point cloud (140) from a first pose to an unknown second pose for a plurality of relative pose candidates. The processing circuitry is further configured to pair each transformed key point with a target point in the target point cloud (150) based on distance, and to determine the relative pose of the mobile agent based on the plurality of feature vectors and positions of the plurality of transformed key points and target points.

The invention particularly relates to a difficulty adaptive reinforcement learning method, device and equipment for a mobile agent, and the method comprises the steps: determining a sampling range in an interaction environment, and carrying out the multi-round sampling in the interaction environment based on the sampling range, and obtaining a multi-round sampling track; on the basis of a difficulty self-adaptive positive sample playback mechanism, calculating a dominant score of each round of sampling trajectory, sorting multiple rounds of sampling trajectories according to the dominant scores, screening trajectories corresponding to the dominant scores meeting a preset buffer condition from a sorting result to obtain multiple successful trajectories, and forming a high-quality playback pool on the basis of the multiple successful trajectories; and extracting a key track in the high-quality playback pool, and updating a strategy network of the mobile agent by utilizing the key track to mix with the online data according to a preset proportion, so as to optimize a pre-established reasoning warm start model according to the updated strategy network. Therefore, the problems of low exploration efficiency, poor generalization and the like of the mobile intelligent agent in an interaction environment are solved, and efficient generalization is realized.

The application discloses a multi-level feature multicast transmission method and device, equipment, storage medium and product, and relates to the technical field of semantic communication. The method comprises the following steps: acquiring multi-level features according to a target picture, the multi-level features comprising semantic features; determining semantic errors of different levels according to the semantic features; and determining the communication resources of data receiving equipment in the current level according to target features of the target picture. The receiving capabilities of different users are acquired, so that reasonable resource allocation can be performed when subsequent semantic communication is performed for multiple users. By establishing an optimization problem based on the communication resources of the data receiving equipment of different levels and the semantic errors of different levels, an optimal solution is obtained. Furthermore, the target features are multicast transmitted based on the optimal solution, the communication resource overhead in the multi-user scenario is reduced, the requirements of different mobile agents are met in information transmission, and the transmission delay and resource consumption are reduced as a whole.

The invention relates to the technical field of passive optical network (PON), is applied to fiber to room (FTTR), enterprise-level fiber to room-Bus (FTTR-B) and broadband fusion terminal products, and particularly relates to an OUN gateway optimization method, device, equipment and medium based on a PON network, by obtaining a spatial radio frequency map and a resource availability state of a mobile agent node in a target area, the OUN gateway optimization method, device and equipment based on the PON network can be used for optimizing an OUN gateway, and the OUN gateway optimization method, device and medium based on the PON network can be used for optimizing the OUN gateway according to the spatial radio frequency map and the resource availability state of the mobile agent node in the target area. And determining whether to trigger a network coverage optimization process according to the first network quality parameter and the resource availability state and the first network quality parameter and the real-time space coordinate of the target terminal device, if so, determining an optimal moving path of the mobile agent node based on the spatial radio frequency map and the real-time space coordinate, and if not, determining the optimal moving path of the mobile agent node. And the mobile agent node is controlled to move to the target deployment position according to the optimal moving path, and the network access connection is completed by using the preset protocol, so that the intelligent path planning and dynamic network optimization of the mobile agent node are realized, and the network coverage can actively adapt to the actual position and demand of the user.

The present invention belongs to the field of mobile communication technology and discloses an airborne federated learning system and a learning method thereof that integrates synaesthesia and computing. The learning system includes a model learning module, a device perception module, a mobile agent, and a core server. The learning method is as follows: the mobile agent first trains a local model based on regional representative weights and then uploads the local model. The core server perceives the location and status information of the mobile agent, updates the regional representative weights, and sends the corresponding power compensation coefficients to all mobile agents. The core server then aggregates the global model and broadcasts the global model for the next round of federated learning. In non-independent and identically distributed scenarios, the present invention considers the location information and channel status of all mobile agents and updates the global model based on local gradient deviation to reduce the impact of local sample deviation.

The application discloses a heterogeneous mobile intelligent agent cooperative control and evolution method, which comprises embedding an emotion cognition module in an artificial mobile intelligent agent, acquiring emotion clues of different granularities to construct evolution rules of the artificial mobile intelligent agent, including acceleration rules, deceleration rules, random slowing rules of introducing a random slowing probability function and position updating rules; constructing automatic mobile intelligent agent evolution rules based on an improved interaction potential energy field function, including acceleration rules, deceleration rules and position updating rules; the acceleration rules are used for adjusting speed; the deceleration rules are used for avoiding collision; the random slowing rules are used for random deceleration of the artificial mobile intelligent agent; and the position updating rules are used for determining position; and the motion path of the heterogeneous mobile intelligent agent is controlled and planned through a field force planning method. The application solves the cooperative control problem of the heterogeneous mobile intelligent agent in a complex dynamic environment, enhances the cooperative ability between the heterogeneous intelligent agents and improves collision avoidance safety.

The utility model relates to the technical field of intelligent agent charging and discharging, and discloses a power supply system and a mobile intelligent agent, which comprise a battery, an automatic charging protection unit, a manual charging protection unit, a discharging protection unit, a control unit and an infrared sensing unit, the output end of the automatic charging protection unit is connected with the output end of the manual charging protection unit, the charging end of the battery and the power supply end of the control unit; the input end of the manual charging protection unit inputs a second power supply voltage; the input end of the discharge protection unit is connected with the discharge end of the battery, and the output end of the discharge protection unit is connected with the power supply end of the control unit and outputs third power supply voltage; the input end of the infrared sensing unit inputs an external infrared signal, and the output end of the infrared sensing unit is connected with the input end of the control unit. According to the utility model, charging and discharging protection and infrared positioning functions are integrated, and the power supply reliability can be improved.

The invention discloses a heterogeneous mobile agent cooperative control and evolution method, which comprises the steps of embedding an emotion cognition module in an artificial mobile agent, and obtaining emotion clues with different granularities to construct an evolution rule of the artificial mobile agent, comprising an acceleration rule, a deceleration rule, a random moderation rule in which a random moderation probability function is introduced, and a position updating rule; constructing automatic mobile agent evolution rules based on the improved interaction potential energy field function, wherein the automatic mobile agent evolution rules comprise an acceleration rule, a deceleration rule and a position updating rule; the acceleration rule is used for adjusting speed; the deceleration rule is used for avoiding collision; the random slowing-down rule is used for randomly slowing down the manual moving agent; the position updating rule is used for determining positions; and the motion path of the heterogeneous mobile agent is controlled and planned through a field force planning method. According to the method, the problem of cooperative control of the heterogeneous mobile intelligent agents in a complex dynamic environment is solved, the cooperative capability among the heterogeneous intelligent agents is enhanced, and the collision avoidance safety is improved.

The present invention discloses a mobile terminal intelligent body information processing system based on prior rules and multimodal large models, which relates to the field of artificial intelligence technology, including an icon accurate positioning and recognition module, which accurately positions and recognizes icons in the interface through different levels of recognition methods, and infers the functions of the icons; an execution efficiency improvement module, which improves the execution speed by task disassembly and optimization path design; an operation optimization module for specific applications, which is used to record the position information, size and corresponding icon function of each interface element, build a navigation path library for specific applications, and optimize the execution path of periodic tasks; a navigation and task data collection and update module, which is responsible for collecting and updating the dynamic data of navigation paths and task execution, and continuously performing self-learning and optimization. Therefore, the above method can achieve accurate icon positioning and function recognition, improve execution efficiency and reduce computing resource consumption, and at the same time be able to cope with interface changes and continuously optimize.

The invention relates to a distributed cross-domain network maintenance positioning system and method based on a mobile Agent, and the system comprises a central network manager which is provided with a central side NMS, and is used for carrying out the configuration of a mobile test node according to a fault service link, and carrying out the positioning of a network problem; a plurality of sub-domain networks, wherein each sub-domain network is used for realizing a corresponding service; the sub-domain network comprises: a service node; the sub-domain side NMS is used for reporting a fault service link and pushing the configured mobile test node to a service node to perform connectivity mutual access of a simulation task; the firewall is used for carrying out session monitoring on the service flow reaching the sub-domain network; and the center side NMS carries out network problem positioning according to the session monitoring structure of the protection wall when the mobile test node carries out the connectivity mutual access of the simulation task. According to the invention, a distributed network management architecture is combined with a mobile agent mode, so that network connectivity testing, problem collection, solution strategy one-key issuing and problem verification whole-process problem closed loop can be conveniently and accurately realized.

The present invention relates to a method for operating a mobile agent (1) based on the recognition of fast-movable environmental objects, in particular for performing a collision avoidance function, having the following steps: detecting (S1) the positions of one or more movable objects (2) in the environment (U) of the mobile agent (1); assigning (S2) the detected environmental objects (2) to a cell grid according to their positions, the cell grid corresponding to a 2D polar coordinate grid having a cell system extending annularly around the mobile agent (1), wherein the number of cells (Z) in the radial direction decreases as the distance from the mobile agent (1) increases; propagating (S2) the environmental objects assigned to the cells (Z) via the cells (Z) in order to obtain radial velocity data, in particular average radial velocity, for each cell (Z); signaling (S5) the collision probability for each cell (Z) according to the radial velocity data (v rad ).

A computer-implemented method for operating a mobile agent in an environment on the basis of a pose of the mobile agent. The location of the mobile agent is determined by way of the following steps in order to operate the mobile agent: capturing sensor data regarding walls and / or objects located in an environment, wherein the sensor data indicate the orientation and distance of the walls and / or objects in an agent-based agent coordinate system; establishing the pose of the mobile agent in the environment using a SLAM algorithm while taking account of a Manhattan orientation.

The invention discloses a multi-level feature multicast transmission method and device, equipment, a storage medium and a product, and relates to the technical field of semantic communication.The method comprises the steps that multi-level features are obtained according to a target picture, the multi-level features comprise semantic features, semantic errors of different levels are determined according to the semantic features, and the target picture is obtained; and determining the communication resource of the data receiving device in the current hierarchy according to the target feature of the target picture, thereby obtaining the receiving capabilities of different users, and facilitating reasonable resource allocation during subsequent semantic communication for multiple users. An optimization problem is established based on communication resources of different levels of data receiving equipment and semantic errors of different levels to obtain an optimal solution, and furthermore, multicast transmission is performed on target features based on the optimal solution, so that the communication resource overhead in a multi-user scene is reduced, the requirements of different mobile agents are met in information transmission, and the information transmission efficiency is improved. And the transmission time delay and the resource consumption are reduced on the whole.