56 results about "Motion strategy" patented technology

The invention discloses a hybrid navigation method, device and system based on laser radar and light reflection boards. The method comprises the following steps: acquiring point cloud data of the laser radar; calculating pose information of the carrier according to the point cloud data of the laser radar; and formulating a motion strategy, and driving the carrier to move, wherein the step of calculating the pose information of the carrier according to the point cloud data of the laser radar comprises the steps of judging whether the point cloud data of the laser radar includes at least three special data accordant with return signal features of the light reflection boards, if so, calculating pose information of the carrier according to the special data, otherwise, directly calculating poseinformation of the carrier according to the point cloud data. According to the method, the laser radar is cooperated with the light reflection boards in use, the accurate position of the carrier canbe calculated according to the special data with relatively high strength returned by the light reflection boards, the position can be further corrected when the environmental scene is changed hereby,and the light reflection boards can be flexibly distributed according to requirement, so that the distribution density of the light reflection boards is decreased, and a dead corner on which the carrier is lost is not formed.

The embodiments of the invention disclose an unmanned aerial vehicle flight obstacle avoidance method, an unmanned aerial vehicle flight obstacle avoidance device, an electronic device and a storage medium. The method includes the following steps that: point cloud observation data collected by an unmanned aerial vehicle are acquired, wherein the point cloud observation data comprise spatial pointcloud data and the attitude information of the unmanned aerial vehicle corresponding to the acquisition time point of the spatial point cloud data; motion trajectories under preset selectable motion strategies are predicted according to the current attitude information of the unmanned aerial vehicle and the selectable motion strategies; probabilities that obstacles exist in each predicted motion trajectory are calculated according to the point cloud observation data and the motion trajectories; and a target motion strategy is selected according to the probabilities that the obstacles exist ineach predicted motion trajectory so as to be adopted to control the movement of the unmanned aerial vehicle. With the unmanned aerial vehicle flight obstacle avoidance method, the unmanned aerial vehicle flight obstacle avoidance device, the electronic device and the storage medium of the invention adopted, the movement of the unmanned aerial vehicle can be predicted in real time in a flying process, and probabilities that the obstacles exist in the predicted paths are calculated; and the movement strategies are further adjusted according to the probabilities of the existence of the obstacles,so that the unmanned aerial vehicle can fly smoothly at a high speed when avoiding obstacles.

The invention discloses a mobile robot visual navigation method based on deep learning, which includes: a mobile robot performing image (color images) acquisition in an unknown environment space through its in-built depth camera, processing the collected images to obtain a preset size picture, marking the picture with a marking tool (such as LabelImg), training the color images with a deep learning method to obtain a target detection model, and inputting the target. The robot explores the unknown environment and collects real time images (including color images and depth images). The color image is used as the input of the target detection model to detect the target. After detecting the target, the direction and the distance of the target position to the robot are calculated according to the depth map to generate a motion strategy.

The invention provides a deep reinforcement learning-based complex terrain self-adaptive motion control method for a hexapod robot. Compared with a conventional mode of realizing motion control of therobot via beforehand programming, the method disclosed in the invention is advantageous in that deep reinforcement learning is used to allow the robot to adaptively adjust motion strategies accordingto complex changes of environment, and "survival rates" and adaptability in complex environments can be improved; compared with a single actor-critic deep reinforcement learning framework, the methodis advantageous in that a multiple actor-multiple critic deep reinforcement learning framework is realized, and training can be sped up and a neural network can be converged fast during deep neural network training operation.

The invention discloses a fully-automatic calibration method for a hand-eye robot based on an exponential product model and particularly relates to a fully-automatic calibration method for achieving synchronization of three parts including a robot body, hands and eyes as well as a camera. The method can be achieved by utilizing a mirror and a plurality of beacon scaffolds instead of additional auxiliary measuring instruments. The fully-automatic calibration method is mainly characterized in that firstly, hand information outside the view field of a terminal camera is converted to a coordinate system of the camera by utilizing mirror reflection principle to finish hand and eye calibration; and secondly, the terminal camera carried by the robot is used as a measuring instrument and a checkerboard on a spacial mirror surface is used as a measuring target, so that the times for shooting the target by using the camera in a rotating process of each joint of the robot are increased, and meanwhile, the camera calibration and the joint parameter calibration are achieved under a special motion strategy. The fully-automatic calibration method for the robot is simple and effective.

The invention discloses a motion detection system, a motion detection terminal and a cloud platform. The motion detection terminal is used for acquiring and storing feature data and / or physiological data of a user, and computing the matching degree with a stored motion strategy according to the feature data and / or the physiological data, comparing the physiological data with the motion strategy with the highest matching degree so as to evaluate whether the motion of the user is normal, wherein the feature data comprise gender, age or health state, the physiological data comprises an acceleration parameter, a breath parameter or a heart rate parameter; the cloud platform is used for storing the motion strategy provided by the motion detection terminal. The system and terminal disclosed by the invention are used for providing an ideal pace strategy for the user under the condition of different motion modes and different physical states; the pace strategy is safe, efficient, reliable and dynamic, and the step of computing the pace strategy is simplified.

The invention discloses a remote diagnosis method for the motion behavior of a protection element at a dispatching terminal. The remote diagnosis method is characterized in that the method comprises following steps: S01: establishing an association model of primary equipment and relay protection; S02: transmitting and collecting relay protection intermediate logic node information, SOE information, protection motion information, telecommand deflection information, and fault record files from a substation to the dispatching terminal; S03: determining whether a power grid has faults; S04: determining a fault point via protection motion information, breaker deflection, and protection fault range-measuring information and obtaining association relay protection if the power grid has faults; S05: extracting intermediate logic node files of association relay protection; S06: comparing and analyzing each protection member according to self-consistency and consistency of each protection member and motion strategy; and S07: determining whether all the protection member motions are correct. According to the method, hidden problems of the protection members are discovered in advance, and necessary data is provided for remote operation and maintenance of relay protection.

The invention discloses a mechanical arm motion planning method based on deep reinforcement learning. The method comprises the following steps of 1, acquiring an environment image once before a mechanical arm moves, wherein the environment image comprises the mechanical arm in an initial state, a moving target point and an intermediate obstacle; 2, according to the acquired environment image, utilizing the target partitioning algorithm to separate a prohibited area, a working area and a target position from one another, and reconstructing a planning space; 3, dividing the reconstructed planning space into three-dimensional grid spaces, and establishing binarized grid spaces; 4, solving the corresponding analytical solution of each joint of the mechanical arm under known terminal coordinates by utilizing the robot inverse kinematics, and determining the relative positional relationships between the mechanical arm and the planned space boundary, the prohibited area boundary and the moving target under the global coordinate system; and 5, planning the motion strategy for the mechanical arm and acquiring the optimal motion strategy, so that the mechanical arm moves to the target position at the minimum cost under the premise of avoiding the obstacle.

The invention discloses a mobile robot navigation method. An image acquisition device is arranged overhead in a navigation area. The method comprises the following steps: step one, collecting a mobilerobot environment image once before the robot motion; step two, acquiring environmental obstacle information according to the collected environment image; step three, establishing a binary raster mapbased on the obtained environmental obstacle information and marking a passable area and non-passage area on the raster map; step four, establishing a mobile robot running rule based on the raster map; step five, setting a total round number M and a shallow trial learning round number M1; to be specific, carrying out shallow trial learning based on a mobile robot running rule to obtain a preliminary Q table; and updating the Q table by reinforced learning based on an initial position p0 of the mobile robot; an step six, according to the updated Q table, acquiring an optimal motion strategy pai<*> of the mobile robot to obtain a motion path of the mobile robot. Therefore, the invalid exploration of the robot in the training process is reduced; the learning efficiency is high and the convergence is fast.

The invention discloses a method and device for determining movement strategy of an unmanned vehicle. The unmanned vehicle collects images continuously in the driving process, and thus, an image collected at present time can be determined, the image is input to the coding end of a pre-trained decision model to obtain a road condition characteristic vector corresponding to the present time, and road condition characteristic vectors, obtained by the coding end of the pre-trained decision model historically, corresponding to different historical time, the road condition characteristic vector corresponding to the present time and a motion strategy of the unmanned vehicle at the present time are input to the decoding end of the decision model, and a motion strategy of next time of the unmannedvehicle is obtained.

The application discloses an automatic walking device comprising a control processing device disposed inside the automatic walking device, a visual image sensor and a multi-line laser radar. The visual image sensor and the multi-line laser radar are arranged at one side of the marching direction of the automatic walking device. The multi-line laser radar emits multiple detecting laser beams basedon a preset angle and determines characteristic information of an obstacle based on reflected signals of the detecting laser beams; the visual image sensor is used for acquiring image information of the marching direction of the automatic walking device. A control processing device calculates and performs the motion strategy of the automatic walking device according to the obstacle characteristicinformation obtained by the multi-line laser radar and the image information, obtained by the visual image sensor, of the marching direction. The application also discloses a robot. Therefore, obstacles with certain heights can be detected; the small-sized obstacles are not neglected; the blind area occurrence is avoided in the detection process; the installation of the camera is reduced, therebyreducing the algorithm difficulty and the manufacturing cost.

Motion strategies in two and three dimensions for scanning microscope imaging are described. An object, sample, or specimen is mounted on a precision three-dimensional stage. The object is moved concurrently with respect to a first axis and a second axis orthogonal to the first against a cutting tool to cut the object. An image of the cut portion is generated as the object is moved. The cutting tool may act as an optical waveguide for illuminating the portion of the object cut. An optical element captures images of the cut and illuminated object. The object may further be concurrently moved with respect to a third axis orthogonal to both the first and second.

The invention provides a multi-robot motion planning method and system and a storage medium. The method comprises the following steps: acquiring the motion state and environment information of robots in motion; determining a motion strategy of the robot through reinforcement learning according to the discretized motion state and environment information, wherein the continuous action in the motion strategy is obtained by determining a reinforcement learning state space by using a fuzzy neural network and outputting the obtained continuous action; determining a basic behavior of the robot according to the motion strategy, and performing cluster motion, wherein the basic behaviors comprise at least one of the following behaviors: advancing to a target, obstacle avoidance movement, collision avoidance movement and movement along a wall. The method solves the problems that the convergence speed is too low and the predictive ability is poor when optimal behavior strategy learning is carried out on the intelligent body in a huge state space and a dynamic change environment, and can be widely applied to the technical field of robot control.

The invention belongs to the technical field of system management, and particularly relates to a method for setting motion strategy according to an assigned target. In the method provided by the invention, the user can detail the own target strategy only by using the motion strategy of other people for reference, which is different from the existing method for detailing the strategy by utilizing modes and reasoning. The method comprises steps as follows: (1) other users sets different motion strategy sets for similar targets, and the motion strategy sets are stored in a strategy library; (2) a method user selects a target and sets detailing parameters to obtain the corresponding motion strategy; and (3) the method user finely adjusts the obtained motion strategy, thereby detailing the target strategy.

The invention relates to a DC power transmission system traveling wave protection method and belongs to the VHV DC transmission technology field. According to the method, a line current and a line voltage of a local station are acquired, polar waves PI are figured out, S conversion on the polar waves is carried out for analysis, time frequency distribution characteristics of polar wave S conversion module values are acquired, and time of the largest abrupt change is the fault generation time; processing is carried out according to protection criteria and protection motion strategies, the criteria are that Pset is greater than or equal to Krel*Pmax, the reliable coefficient Krel is greater than 1, the Pmax is the largest value of polar wave S conversion module values of all outer-zone faults and disturbance, moreover, a polar wave setting value acquired through the formula above is obviously smaller than a first peak value of inner-zone fault polar wave S conversion module values, so no error protection motion generation is not only guaranteed during outer-zone faults, but also inner-zone protection motion is guaranteed to be reliable. The DC power transmission system traveling wave protection method can rapidly and accurately determine the faults and avoids line traveling wave protection error motion.

An autoloader system with modularization architecture and self-adaptive motion control ability for mass optical disks duplication includes four physical modules: a robot arm module with sensors and joints dedicates for delivering and picking up optical disks; an optical disk duplication tower has a number of optical drives in a stack, or a matrix for optical disk duplication; a motion control module has an embedded motion controller and a power source to synchronize the motion of robot arm and duplication; a platform module has a base frame to fix other modules and a user interface. Some disk stacks are situated on top of platform module. The use of a self-adaptive control algorithm, consisting of a Motion Strategy Database, Initial Process, Motion Planning Process, Motion Generation Process and Motion Monitor Process, to ascertain system configurations and components furthest satisfy the required flexibility for modifying / upgrading hardware or ever-changing user needs. The use of DC motors and self-correcting adaptive algorithm provides better versatility to disk copier systems than most commonly used stepper motors, even in the case of “short tray” wherein the tray does not fully extend out. The stand-alone design of present invention further makes the use and operation of the disk copier easier and friendlier.

A method for predicting the trajectories of extraneous objects in the surroundings of an ego vehicle, and for determining a separate future trajectory adapted thereto for the ego vehicle. The method includes identifying the extraneous objects are identified. It is ascertained toward which proximate destination the movement of each of the extraneous objects is headed and according to which basic rules this movement occurs. It is ascertained toward which proximate destination the movement of the ego vehicle is headed and according to which basic rules this movement occurs. One quality function each is established for the ego vehicle and the extraneous objects. One quality measure each is established for the ego vehicle as well as for the extraneous objects. Those optimal movement strategies of the ego vehicle and of the extraneous objects that maximize the quality measure are ascertained. The sought trajectories are ascertained from the optimal movement strategies.

The embodiment of the invention relates to the technical field of intelligent processing, and particularly discloses a method and device for intelligently providing a motion strategy, and intelligentwearable equipment, and the method comprises the steps: obtaining the motion amount consumption of a user in a unit time period; comparing the exercise amount consumption in the unit time period withthe pre-obtained standard exercise amount consumption to obtain a comparison result; and providing a motion strategy in the next unit time period for the user according to the comparison result. Through the method, blind movement of the user can be avoided. After the collected exercise amount of the user is consumed, the exercise amount is no longer simply provided for the user to check statically, but is compared with the standard exercise amount consumption to determine whether the current exercise amount consumption is reasonable or not, and an exercise strategy of the user in the next unittime period is given according to whether the current exercise amount consumption is reasonable or not. And the physical health of the user is ensured, and unnecessary harm to the body caused by overwork exercise of the user can be avoided.

The invention discloses a cleaning equipment motion control method which comprises the steps: detecting whether cleaning equipment is collided or not, if the equipment is collided and the cleaning equipment is driven to rotate, continuously irradiating and receiving a detection signal of an obstacle, and stopping spinning when the value of the detection signal is reduced; calculating to obtain anobstacle color parameter according to the received detection signal; adjusting and executing a preset edge movement strategy according to the obtained obstacle color parameters; and adjusting and executing a preset edge movement strategy according to the obtained obstacle color parameters. According to the cleaning equipment motion control method, the current obstacle color coefficient is obtainedby detecting the obstacle signal values at intervals, the obstacle color is judged according to the obstacle color coefficient, and the edge motion strategy corresponding to the obstacle color is executed; therefore, the situation that existing cleaning equipment cannot detect black obstacles and glass obstacles is avoided, the probability of collision is avoided, and the sweeping efficiency is improved.

The invention discloses a high-precision three-dimensional motion track acquisition and motion process reproduction method based on a binocular video sensor. The method is mainly applied to athlete movement condition data acquisition and comprehensive analysis of racing events such as athletes' athletic tracks, swimming and ice competitions, helps athlete coaches to effectively analyze movement characteristics, improves movement strategies and improves movement performance. The method mainly comprises the following two aspects: firstly, a high-precision video target tracking technology and a servo control tracking system are adopted to determine the angle and orientation of a target; secondly, the tracked pixel coordinate data and turntable corner data are converted into three-dimensionalspace coordinates through a camera calibration and binocular stereoscopic vision technology, and then the motion trail of the moving target in the three-dimensional space is obtained; and then opticalprojection can be carried out on a certain space plane such as a sports field plane through a high-precision servo control system, so that the motion process of an original target is reproduced, andthe projection position, speed, acceleration and the like at each moment are consistent with the previously acquired motion process of the target.

The invention discloses a robot welt motion control system and method. The system comprises a first sensing unit which is used for horizontally collecting the front point cloud data of a robot and thedistance value of a wall or a nearest obstacle, a second sensing unit which is used for collecting a distance value of a right side wall or a nearest obstacle of the robot, a third sensing unit whichis used for vertically acquiring 45-degree point cloud data in front of the right of the robot and a nearest obstacle distance value, and an industrial personal computer which is used for sending a control instruction according to the data acquired by the first sensing unit, the second sensing unit and the third sensing unit, so that the robot moves along the corridor welt according to a set movement strategy and finally returns to and stops at a starting point position. According to the invention, an operator does not need to enter a dangerous area for manual mapping or priori data acquisition, so that the labor intensity and the operation risk are reduced; and the system has high adaptability to the environment, is very suitable for autonomous operation in dangerous environments such asinfectious disease hospitals and the like, and has high commercial popularization and market application values.

The invention discloses a centralized relay protection device. The centralized relay protection device comprises far-end data collecting units, far-end motion executing units, a network switch unit and a centralized control unit. The far-end data collecting units are used for collecting information of an electric power system and transmitting the information to the centralized control unit through the network switch unit, the centralized control unit generates a motion set, and the motion set is issued to the far-end motion executing units through the network switch unit. Control devices are deployed in a centralized mode, collecting and motion devices are deployed in a dispersed mode, and based on synthesizing positions of all the collecting and motion devices in electric power topology and historical operation data information of a whole net, the optimal motion behavior is done through decision making conveniently; an electric power topology network and a relay protective motion strategy can be configured on the centralized control devices flexibly, and the maintainability and the expandability of the system are improved.

The invention discloses a method and a device for training a model and determining a motion strategy of an unmanned vehicle, and the method comprises the steps: determining a plurality of groups of continuous two-frame images in each continuous frame image collected in history as each training sample, determining a bounding box of a second image as a label for each training sample, and through a preprocessing model, determining a pre-processing result, determining a prediction result and a post-processing result of a first image of a training sample through a detection model, determining a detection result through a classification layer, and training a to-be-trained detection model by taking the minimum difference between the detection result and a label as an optimization target. According to the method, the obstacle instances of the first image in the training samples are predicted, at least part of the to-be-selected bounding boxes of the obstacle instances of the second image are updated, finally the bounding boxes of the obstacle instances of the second image are determined through the classification layer and serve as the detection result, manual parameter adjustment is not needed, and the accuracy is higher.

The invention provides a camera holder control method and device, equipment and a storage medium, relates to the technical field of security and protection, and aims to improve the monitoring effect of a camera. The method comprises the following steps: when a first monitoring target enters a monitoring area of a camera, determining an initial motion strategy adopted by a camera holder for monitoring the monitoring target based on the equipment structure information of the camera holder; controlling the camera holder to rotate based on the initial motion strategy; in the execution process according to the initial motion strategy, determining the motion stage of the camera holder during out-of-step according to the actual rotation angle of the camera holder at each moment and the estimated rotation angle of the camera holder at the corresponding moment determined based on the initial motion strategy; adjusting the initial motion strategy based on the motion stage of the camera holder during out-of-step, and obtaining an adjusted motion strategy; and when the monitoring target of the same type as the first monitoring target is monitored, controlling the camera holder to rotate based on the adjusted motion strategy.

The invention discloses a speed control method and device for unmanned equipment, which can be applied to the unmanned equipment in the unmanned field, such as an unmanned vehicle. The method comprises the steps: determining a stable passing speed corresponding to an uneven road surface according to the cargo integrity after the cargo loaded by test equipment passes through the uneven road surface; when the driving path of the unmanned equipment has an uneven road surface, determining nursing between the unmanned equipment and the uneven road surface, so as to determine a motion strategy of the unmanned equipment according to the current speed, the stable passing speed and the distance between the unmanned equipment and the uneven road surface, and according to the motion strategy, controlling the unmanned equipment to pass through the uneven road surface at the stable passing speed. According to the motion strategy of the unmanned equipment determined by the method, the influence of an uneven road surface area on the integrity of an object loaded by the unmanned equipment in the driving process is considered, and the completeness of a distribution task is ensured.

The invention discloses a line bus integrated backup protection method using a pilot comparison principle based on the fault direction. The method comprises the steps: taking a breaker as the center, determining the electrical elements related with the breaker and the backup protection functions, and incorporating the backup protection functions in the same protector to complete; determining the fault direction information is interacted among which protectors, and the mode for interaction of the fault direction information; comparing the fault direction information when a fault occurs, and determining the position wherein the fault is according to the compared result of the fault direction information; and determining the motion strategy for backup protection. The line bus integrated backup protection method using a pilot comparison principle based on the fault direction has the advantages of reducing the quantity of backup protection, simplifying the coordination relationship among backup protection, simplifying the control loop of the breaker, improving the reliability for working of a secondary system, being able to determining the fault position rapidly and reliably when the fault occurs, and improving the rapidity and the reliability of the backup protection motion.

The embodiments of the invention disclose an unmanned aerial vehicle flight obstacle avoidance method, an unmanned aerial vehicle flight obstacle avoidance device, an electronic device and a storage medium. The method includes the following steps that: point cloud observation data collected by an unmanned aerial vehicle are acquired, wherein the point cloud observation data comprise spatial pointcloud data and the attitude information of the unmanned aerial vehicle corresponding to the acquisition time point of the spatial point cloud data; motion trajectories under preset selectable motion strategies are predicted according to the current attitude information of the unmanned aerial vehicle and the selectable motion strategies; probabilities that obstacles exist in each predicted motion trajectory are calculated according to the point cloud observation data and the motion trajectories; and a target motion strategy is selected according to the probabilities that the obstacles exist ineach predicted motion trajectory so as to be adopted to control the movement of the unmanned aerial vehicle. With the unmanned aerial vehicle flight obstacle avoidance method, the unmanned aerial vehicle flight obstacle avoidance device, the electronic device and the storage medium of the invention adopted, the movement of the unmanned aerial vehicle can be predicted in real time in a flying process, and probabilities that the obstacles exist in the predicted paths are calculated; and the movement strategies are further adjusted according to the probabilities of the existence of the obstacles,so that the unmanned aerial vehicle can fly smoothly at a high speed when avoiding obstacles.