35results about How to "Improve task completion rate" patented technology

The invention provides a servo system for instructing a pan-tilt system of an unmanned aerial vehicle (UAV) by adopting head movement of an operator, belonging to the field of UAV control. The servo system is characterized in that the servo system comprises a head attitude sensor, a ground station, data radios, video radios and a UAV pan-tilt system capable of pitching and yawing, wherein the head attitude sensor is used for detecting the head movement of the operator; the ground station is used for computing the deflection command according to the head movement and sending the deflection command to the vehicle-mounted pan-tilt system via the data radios to ensure the vehicle-mounted pan-tilt system and a camera carried by the vehicle-mounted pan-tilt system to deflect towards the direction expected by the operator; the camera returns the shot videos to the ground station via the video radios and the ground station displays the shot videos in real time; and the operator corrects the deflection rate and direction of the vehicle-mounted pan-tilt system in real time according to the videos displayed by the ground station to ensure the camera to aim at the expected target all the time. By using the servo system, the operator can control deflection of the vehicle-mounted pan-tilt system by adopting the gut reaction of own head movement, and the servo system has the characteristics of being easy in operation and convenient use in and can obviously reduce the workload of the UAV operator.

The invention belongs to the technical field of vehicle-mounted communication, and discloses an Internet of Vehicles content edge unloading method and a mobile resource allocation system. Resource information of a vehicle node is managed, tasks are divided, a target node is combined, and thus, calculation unloading in a vehicular Ad Hoc network is realized; a calculation unloading target node selection method is adopted, in reference to the calculation ability and the time capable of providing service of a node, a corresponding task amount is allocated to the node, and according to the information of each resource, a vehicle is screened. The calculation unloading problem in an interconnection scene between high-mobility vehicles can be solved, and the target of performing calculation at any time and at any place by a user is realized. The resource information of vehicle nodes with high mobility is managed, task dividing and target node selection are combined, and calculation unloadingin the vehicular Ad Hoc network can be realized; and on the basis of comprehensively considering multiple factors, the calculation unloading target node selection method is adopted, and the vehicle isscreened according to the information of each resource.

The invention discloses a path planning method based on multi-agent enhanced learning, and belongs to the technical field of aircrafts. Firstly, a global state division model of an air flight environment is established, and a global state transfer table Q-Table 1 is initialized, and the global state of a certain line is randomly selected to be used as the initial state s1; in all columns of the current sate s1, an epsilon-greedy algorithm is adopted to select a certain column to be recorded as a behavior a1; based on the selected behavior a1, the next state, a formula which is as shown in thespecification, of the current state s1 is obtained in the global state transfer table Q-Table 1; the specific element values corresponding to the current state s1 and the behavior a1 are updated in the global state transfer table Q-Table 1 according to a transfer rule of a Q-Learning algorithm; a formula which is as shown in the specification is updated to enter an inner layer cycle; and a local planning path corresponding to the updated state s1 is obtained by adopting the Q-Learning algorithm. The number of iterations of the outer-layer cycle is increased by 1 until N1, and global path planning of the aircraft in the air is completed. The aircraft can meet requirements of different environments, so that the survival rate and the task completion rate of the aircraft are improved, and theconvergence speed of the enhanced learning is improved.

In the measurement system for an aircraft according to the present invention, the three-axis components of the airspeed of the aircraft, the three-axis components of the wind speed currently encountered by the aircraft, and the wind speed distribution ahead of the aircraft, are measured simultaneously, by providing a Doppler anemometer which measures the airflow speeds at a plurality of points separated by a prescribed distance in the beam irradiation direction, and an inertial data measurement device which measures the movement information and positional information of the aircraft. Furthermore, in the measurement system for an aircraft according to the present invention, the measurement accuracy of the device is improved by providing means for obtaining information the movement and positional information of the aircraft during measurement, from the inertial data measurement device, and for correcting the measurement values from the Doppler anemometer by means of the information.

The invention belongs to the technical field of program control except digital control, namely, sequence controllers or logic controllers, and discloses a Multi-Agent-based elastic task planning system and method. The method comprises the following steps of: dividing satellite into K types via a clustering algorithm, and classifying the satellites clustered into a type into a sub-Agent; and dividing the sub-Agents in a comprehensive Agent according to different dynamics of tasks, so as to realize matching between resource granularities and resource granularities of task demands to the greatestextent. According to the method and system, favorable matching between task demands and satellite demands can be effectively realized, system structures can be dynamically adjusted according to tasks, residual resources can be effectively utilized, and rapid and efficient planning of multiple satellites and multiple tasks is realized; and the system and method are elastic, are capable of dynamically adjusting dynamic matching between sub-Agent structures and task resources and utilizing residual resource of each Agent, and are more efficient in resource utilization, higher in task completionrate and suitable for different task scenes.

The invention discloses an edge collaborative storage method for the real-time data processing application in a heterogeneous edge environment. The method comprises the following steps of firstly, determining a neighbor server set; then, calculating the storage benefit of the model data corresponding to each application relative to the cloud for all edge servers, generating an initial storage benefit matrix, wherein the storage benefit is mainly reflected in time delay reduction; searching a storage candidate scheme with the maximum storage benefit from the storage benefit matrix; if the corresponding edge server has enough space to accommodate the model data, storing and updating the residual storage space of the edge server and the storage benefits of other model data influenced by the storage operation; otherwise, setting the benefit value to be infinitely small, and not considering the storage candidate scheme any more; and repeating the two steps until the storage spaces of all the edge servers are not enough to accommodate new model data.

The invention discloses a task processing method, a task processing device, a computer readable medium, and an electronic device, and relates to the technical field of computers. The task processing method comprises the following steps: acquiring a task executed by a target user; and according to the information that whether each user completes the corresponding task or not, determining a task similar to the task executed by the target user as a candidate task, and providing the candidate task to the target user so that the target user can execute the task. With the technical scheme, the taskcompleting rate can be effectively improved, and then the expected operation effect can be favorably realized.

The invention discloses a formation and obstacle avoidance control method for unmanned aerial vehicles in an unmanned aerial vehicle system, and belongs to the unmanned aerial vehicle control technology. The method comprises the following steps: S1, carrying out formation of all unmanned aerial vehicles; S2, judging whether an obstacle exists in a set range of the unmanned aerial vehicles or not according to signals collected by the unmanned aerial vehicles at the moment t: if so, switching to step S3, otherwise, switching to step S5; S3, calculating the output of an obstacle avoidance controller of the unmanned aerial vehicles according to the speeds and positions of the unmanned aerial vehicles at the moment t; S4, calculating the speeds and positions of the unmanned aerial vehicles at the moment t+1 according to the output, the speeds and the positions of the unmanned aerial vehicles at the moment t, adjusting the unmanned aerial vehicles at the moment t+1, enabling t to be equal tot+1, and switching to step S2; S5, adjusting the formation, obstacle avoidance and ending methods according to whether the unmanned aerial vehicle system avoids obstacles and whether the unmanned aerial vehicle system reaches the target. The control method provided by the scheme can solve the problems that the unmanned aerial vehicles keep poor obstacle avoidance during formation navigation and cannot be recombined after obstacle avoidance.

The invention discloses a QoE perception service enhancement method for an edge artificial intelligence application, and the method comprises the following steps: A, enabling an IOT terminal to transmit an operation task to a scheduler, and enabling the scheduler to judge whether to transmit an operation request to a local server or upload the operation request to a data center according to the requirements of the operation task; b, establishing a target and a constraint condition of a service enhancement model of perception quality; and C, establishing a two-stage scheduling strategy to reduce the calculation complexity. According to the invention, the defects in the prior art can be solved, and the service experience of users and the task completion rate are improved.

The embodiment of the invention provides a mobile auxiliary edge computing method and device, a medium and equipment, which are applied to the Internet of Things at least comprising a fixed edge nodeand a mobile edge node, and the method comprises the following steps: requesting an edge task from the fixed edge node according to a user request and resource distribution, wherein the edge task is auser request which cannot be responded by a fixed edge node; auctioning the edge task to a plurality of mobile edge nodes according to an auction mechanism, and determining a final pricing strategy according to the income of each mobile edge node and the expenditure of the fixed edge node; in combination with the pricing strategy and pricing of the mobile edge nodes, determining a global allocation strategy of the edge task at each mobile edge node. According to the method, a mobile auxiliary edge computing framework is provided for the first time, and the service quality of the fixed edge node is improved by introducing the mobile edge node.

The invention discloses a loose coupling integrated flight display. Internal modules of the display comprise a GCPU module, a GPM module, a DCPU module, an IOM module, an SPM module, a mother board, aPS module, a display screen and a box body. The PS module supplies a voltage required by normal work of the modules. The DCPU module, the IOM module and the SPM module are connected through a motherboard local bus, thereby carrying out data exchange. The DCPU module is taken as a bus controller. The IOM module and the SPM module are taken as local bus slave devices. The DCPU module and the GCPUmodule carry out the data exchange through an Internet interface. The DCPU module, the IOM module and the SPM module form a data collection processing function unit. The GPM module is mounted on the GCPU module in a form of a sub-card, and a comprehensive displaying processing function unit is formed. Displaying information generated by the unit is displayed on a displaying screen. The data collection processing function unit and the comprehensive displaying processing function unit work independently through adoption of high performance processors and are interconnected through a serial bus,so a loose coupling structure is formed. A redundant system can be formed by double displays.

The invention provides an emergency scheduling method and system for agile satellite resources, and relates to the technical field of emergency task scheduling. By considering the insertion opportunity of the visible time window of the fixed pitch of the ground station in the scheduling algorithm, the appropriate visible time window of the fixed pitch of the ground station is inserted under the condition that the storage constraint of the agile satellite is not met, so that the storage data of the agile satellite is released to the ground station, the imaging data load of the agile satellite is reduced, more task insertion opportunities are provided, and the overall task completion rate is increased. In the algorithm scheduling process, the emergency degree of the emergency task is calculated according to a designed emergency degree heuristic factor of the emergency task, tasks which are large in weight, few in visible time windows of fixed pitching and close to the completion deadline are preferentially arranged, the influence of task arrangement in advance on the insertion opportunity of subsequent tasks is reduced, and the task completion rate and the overall scheme income are improved.

The invention discloses a communication network field maintenance model migration method based on time delay optimization, which comprises the following steps of: firstly, respectively calculating type similarity, name similarity and interpretation information similarity based on similarity service demand analysis, reducing the influence of irrelevant information, improving the service matching accuracy, and improving the service matching efficiency; and finally, calculating according to the weight of each similarity to obtain a comprehensive similarity. Secondly, migration modes under different conditions are selected according to the similarity of the service requirements, model migration is carried out in a direct migration mode, a backbone network freezing mode and a multi-stage migration mode, and the problems that the intelligent operation and maintenance environment of the communication network is complex and the migration efficiency is low are solved; finally, various indexes such as task time delay, edge node resources and task resource requirements are considered, a multi-priority queue model and an improved discrete particle swarm algorithm are used, migration task unloading and maintenance task unloading in a communication network field maintenance scene are completed, and the purposes of minimizing task time delay and improving task completion efficiency are achieved.

The invention relates to an overlapping organization cooperation control method based on a vehicle fog calculation architecture, and the method is characterized in that each vehicle node in a target region generates task information needing cooperative calculation according to the demands, and transmits the task information to a roadside center unit in a corresponding service region, the vehicle nodes in the same service region form an alliance, and the cooperative calculation of the vehicle nodes in the target region is completed. Each vehicle node is added into a plurality of alliances at the same time, so that the alliances are overlapped, each alliance realizes calculation resource sharing under the control and scheduling of the roadside center unit, and the roadside center unit generates an optimal resource scheduling decision according to calculation resources and position information periodically fed back by vehicles. And the alliance performs competition calculation according to the optimal resource scheduling decision. Compared with the prior art, the method has the advantages that the task calculation efficiency of the local Internet of Vehicles is effectively improved, the task is prevented from being transmitted to the remote server side, and the service time delay and the transmission overhead are reduced.

The invention discloses a task distribution method for radio environment map data collection. The radio environment map data collection is carried out through utilization of a mobile crowd sensing technology; and two-phase task distribution algorithms are employed according to information attributes registered in a system by users in a mobile crowd sensing network, namely data types capable of being sensed by the users and quotations of the users for various sensed data types. According to the two-phase task distribution algorithms, a high-dimensional radio environment map data collection task is divided into a plurality of low-dimensional subtasks based on a k-means clustering thought; and the subtasks are distributed to the right number of appropriate sensing users through adoption of an LCBPA algorithm. An experimental result shows that the method has the beneficial effects that through coordinative sensing among the users in the mobile crowd sensing network, the radio environment map data collection difficulty and cost are reduced, a task finishing rate is improved, the total data collection cost is minimized, and the user participating balance is improved.

The invention provides a multi-satellite task scheduling planning method based on an improved teaching optimization method, and relates to the technical field of aerospace, and the method comprises the following steps: carrying out the mathematical modeling of a multi-satellite task scheduling planning problem into a combined optimization problem with constraints; providing an improved teaching optimization method; solving the combined optimization problem with constraints by using an improved teaching optimization method; and constructing initial solutions. The problem that in the prior art, the task completion rate and the convergence speed need to be increased is at least partially solved, and compared with the prior art, the method improves the convergence speed and the solving success rate, and the task completion rate is also increased.

The invention relates to the technical field of converters, in particular to a traction-auxiliary converter for an internal combustion locomotive and a converter cabinet, and aims at solving the technical problems that an existing internal combustion AC transmission locomotive is low in integration level, poor in energy conservation and emission reduction effect and large in occupied locomotive space. The traction-auxiliary converter comprises a rectifier electrically connected with a main generator, the rectifier is connected with an intermediate direct-current loop; a grounding detection module, a manual discharge module, a brake chopper and an inverter group which are connected in parallel are arranged in the intermediate direct-current loop, the brake chopper is provided with a chopperbrake resistor, and the inverter group comprises an auxiliary fixed-frequency inverter, an auxiliary variable-frequency inverter and three traction inverters which are connected in parallel. The converter cabinet comprises two groups of converters, the auxiliary fixed-frequency inverters of the two groups of traction-auxiliary converters are electrically connected through a fixed-frequency expansion loop, and the auxiliary variable-frequency inverters of the two groups of traction-auxiliary converters are electrically connected through a variable-frequency expansion loop.

The invention discloses a crowdsourcing pricing system evaluation method, a device, equipment and a readable storage medium, and the method comprises the steps: extracting original system parameters from a to-be-evaluated crowdsourcing pricing system, and establishing an initial group set according to the original system parameters; performing data optimization on the initial group set through a genetic algorithm to obtain a new group set; determining a corresponding fitness function value according to the new group set; and evaluating the crowdsourcing pricing system to be evaluated accordingto the fitness function value. By means of the mode, the crowdsourcing price of an original user can be adjusted, the crowdsourcing price is given to a new user point, parameters of a crowdsourcing pricing system can be optimized according to crowdsourcing data completed in real time, and the task completion rate is greatly increased.

The invention discloses a formation and obstacle avoidance control method of unmanned aerial vehicles in an unmanned aerial vehicle system, which belongs to the unmanned aerial vehicle control technology. , determine whether there is an obstacle within the set range of the UAV, if so, go to step S3, otherwise go to step S5; S3, according to the speed and position of the UAV at time t, calculate the obstacle avoidance controller of the UAV Output; S4 calculates the speed and position of the UAV at time t+1 according to the output, speed and position of the UAV at time t, and after adjusting the UAV at time t+1, set t=t+1 and enter the step S2; S5 adjust the formation, obstacle avoidance and ending methods according to whether the UAV system avoids obstacles and whether it reaches the goal. The control method provided by this scheme can solve the problems of poor obstacle avoidance and inability to reorganize after obstacle avoidance when UAVs maintain formation navigation.

The invention discloses a pilot anti-vertigo ability training device, which relates to the technical field of pilot teaching equipment. As for the problem of judgment, this device mainly includes a fixed platform. A support mechanism, a detection mechanism, an intercepting mechanism, a controller, a rotating mechanism, and an adjustment mechanism are arranged on one side of the fixed platform. The pilot stands on the support platform and fixes himself, and the motor drives the rotating platform Rotate, and drive the support platform to rotate itself through the meshing between the gears, so that the pilot rotates himself when the whole body rotates. The beneficial effect of the present invention is that by changing the rotational speed and direction, the pilot's vertigo ability can be trained more comprehensively, and it is easy to know whether the pilot can correctly adjust his own condition and judgment ability when he is dizzy through the detection mechanism, so as to facilitate further improvement. Good for pilot training.

The invention provides a processing method, device and unmanned aerial vehicle for the interruption of the remote control link of the unmanned aerial vehicle. Caused by an interference signal; if it is caused by an interference signal, it is judged whether the remote control link data signal is detected when the timing duration is less than or equal to the interference judgment preset duration T5; if yes, continue to perform the flight mission, otherwise, execute the link Interrupt refractory processing, wherein, the preset time period T5 for judging interference is longer than the preset time period T2 for judging whether it is refractory to interrupt. In the present invention, by adding the judgment of the interference signal in the link interruption judgment of the UAV, if it is judged that the link interruption is caused by the interference signal, the state of the UAV is turned into an interference waiting state instead of entering an interruption difficult state. It can prevent the UAV from abandoning the original mission due to recoverable interference, which greatly improves the mission completion rate and practicality of the aircraft.

In an existing coverage optimization method based on a perceived user track in mobile crowd sensing, only the moving position of a perceived user is considered, and other information of the user, such as moving speed and environmental noise, is neglected. In order to solve the problem, the invention provides a coverage optimization method based on multi-modal trajectory data fusion. The method comprises the following steps: firstly, analyzing track data characteristics of a user in mobile crowd sensing, judging the influence of moving speed and environmental noise on the acquisition intention of the sensing user, and establishing a speed grey-scale map and a sound grey-scale map; then, fusing the speed grey-scale map and the sound grey-scale map by using a double-branch convolutional neural network to form a track grey-scale map; and finally, sensing user selection is carried out based on the grey-scale map coverage similarity, so that the coverage rate is maximized.

The invention belongs to the technical field of in-vehicle communication, and discloses a method for unloading the content edge of the Internet of Vehicles, a mobile resource allocation system, management of resource information of vehicle nodes, task division and combination of target nodes, so as to realize calculation unloading in the in-vehicle ad hoc network; Calculate the offloading target node selection method, assign the corresponding task amount to the node with reference to the node's computing capability and the time available to provide services, and screen the vehicles according to the information of each resource. The present invention solves the problem of computing unloading in the scene of interconnection between vehicles with high mobility, and realizes the goal that users can perform computing anytime and anywhere. Manage the resource information of highly mobile vehicle nodes, combine task division and target node selection, realize the calculation offload in the vehicle ad hoc network; give the calculation offload target on the basis of comprehensive consideration of various factors The node selection method filters the vehicles according to the information of each resource.

The invention provides a cooperative computing task allocation method under edge computing, and belongs to the technical field of edge computing. According to the method, sharing of computing resources of the edge computing server and the industrial Internet of Things device is considered, various heterogeneous devices are promoted to participate in cooperation, the problem of task delay is solved, and compared with a method that all tasks are executed on a main server (a non-cooperative computing task allocation method), the method has better performance in the aspect of delay, accelerates the task execution speed and effectively improves the task completion rate and the resource utilization rate of system equipment.

The invention provides a method and device for processing interruption of a remote control link of an unmanned aerial vehicle and a unmanned aerial vehicle. The method comprises the steps: starting timing when the unmanned aerial vehicle enters an interruption waiting state, and judging whether the link interruption of the unmanned aerial vehicle is caused by an interference signal; if so, judging whether a remote control link data signal is detected or not when the timing duration is smaller than or equal to the interference judgment preset duration T5; if so, continuing to execute the flight task , and if not, executing link interruption difficult-to-recover processing, wherein the interference judgment preset duration T5 is larger than the interruption difficult-to-recover judgment preset duration T2. According to the method and device, the judgment of the interference signal is added in the link interruption judgment of the unmanned aerial vehicle, and if the link interruption is judged to be caused by the interference signal, the state of the unmanned aerial vehicle is converted into an interference waiting state instead of entering an interruption difficult-to-recover state, so that the unmanned aerial vehicle is prevented from giving up an original task due to recoverable interference; and the task completion rate and the actual combat performance of the aircraft are greatly improved.

According to the multi-satellite joint planning method and system considering task splitting provided by the invention, the task planning model is constructed. The model covers planning of detachable observation tasks and non-detachable observation tasks at the same time, a staged algorithm for planning of the non-detachable observation tasks and planning of the detachable observation tasks is designed, and various constraint conditions from satellite resource capacity level, task timeliness requirements and the like are fully considered. A large number of numerical experiments prove that through task splitting, the possibility of cooperation among multiple satellites is provided, the task completion rate is increased, multi-satellite joint observation benefits are obviously increased, and finally the observation efficiency of the satellites is greatly improved.