626results about How to "Avoid falling into" patented technology

The invention discloses a mobile robot path planning method based on an improved RRT* algorithm. The method introduces a target biasing strategy into a standard RRT* algorithm so as to reduce the randomness of sampling points; provides an avoidance step length extension method in order that a random tree can reasonably stay away from an obstacle area and avoids falling into a local minimum; and smoothes a path obtained by the improved RRT* algorithm by using a reverse sequence connection method smoothing strategy, so as to reduce the direction-changing operations of the robot and achieve the stable movement of the robot. Compared with an original standard RRT* algorithm, the improved RRT* algorithm has a better planned path and takes less time.

The invention discloses a cascade reservoir optimal operation method based on the adaptive particle swarm optimization algorithm. According to the method, aiming at the defect of the particle swarm method in cascade reservoir optimal operation, fixed initialization improvement is conducted firstly on particle random initialization to enable the algorithm to have the possibility of approaching the optimal value at the beginning, large-scale dead zones do not exist, convergence speed is increased, and the stability of the algorithm is improved; then according to the group cooperation idea and the cluster ecological niche idea, an initialized group is dynamically divided into three subgroups, optimization and parameter selection are conducted on each subgroup in an adaptive mode according to the difference of particles, and in this way, the particle diversity is improved, the information exchange model is changed, and local optimum of the algorithm is avoided. According to the improved algorithm, the function problems of nonlinearity and multiple local minima can be well solved, and an effective and feasible solution is provided for cascade reservoir optimal operation.

The invention discloses a wind power prediction method based on a modified particle swarm optimization BP neural network. The method includes the following steps: 1. encoding weight values and threshold values of a BP neural network as particles, and initializing the particles; 2. computing each particle fitness value with the difference between the result obtained from BP neural network training and an anticipated value as a fitness function; 3. comparing the fitness value of each particle and individual optimal particle to obtain a global optimal particle; 4. updating the speed and position of the particle; 5. determining whether the global particle meets termination conditions, if the global particle meets termination conditions, terminating the computing and outputting an optimal weight threshold value, and if the global particle does not meet termination conditions, back to step 2 and carrying out iterative operation; and 6. Using the optimal weight threshold value that is acquired by step 5 to connect an input layer, a hidden layer and an output layer of the BP neural network, and obtaining the result of wind power prediction on the basis of the result of the BP neural network. The method has fast convergence speed, high precision, and is not easily trapped to local extremum.

The invention provides a robot path planning method integrating an artificial potential field and a logarithm ant colony algorithm. The method comprises the following steps: S1, initializing; S2, establishing a grid map containing obstacle information; S3, establishing a movable grid table of the ants according to the current positions of the ants; S4, calculating an attractive force and a repulsive force received by the position of the current ant in the artificial potential field, establishing an influence function q (t) of the artificial potential field, and calculating a minimum included angle between a resultant force borne by the ant in the artificial potential field and an adjacent grid direction; S5, improving an ant colony algorithm heuristic function eta ij and a pheromone updating strategy; S6, calculating the transition probability density of the improved ant colony algorithm, and updating the tabu table; S7, judging whether path planning exploration is completed or not, ifnot, entering S3, and if yes, entering S8; and S8, performing re-iteration or ending according to the judgment condition. According to the method, the convergence speed of the ant colony algorithm inpath planning is effectively improved, and the situation that the artificial potential field algorithm is prone to falling into local optimum is reduced to a great extent.

The invention belongs to the technical field of unmanned driving, and discloses a target following and dynamic obstacle avoidance control method for a speed difference slip steering vehicle, and the method comprises the steps: building four neural networks through employing a depth determinacy strategy in reinforcement learning; constructing a cost range of the obstacle so as to determine a single-step reward function of the action; determining continuous action output through an actor-critic strategy, and updating network parameters continuously through gradient transmission; and training a network model for following and obstacle avoidance according to the current state. According to the method, the intelligence of vehicle following and obstacle avoidance is improved, and the method canbetter adapt to an unknown environment and well cope with other emergencies. the complexity of establishing a simulation environment in the reinforcement learning training process is reduced. By utilizing a neural network prediction model trained in advance, the position and posture of each step of the target vehicle and the obstacle can be obtained according to the initial position and posture ofthe target and the obstacle and the action value of each step, so that the simulation accuracy and efficiency are improved.

Provided is a route planning method of an indoor blind guiding robot under a dynamic environment. According to the method, through generation of environmental raster maps, expansion treatment, targetpoint determination, map information collection, planning of a global route and regional routes, barrier avoidance and a mode of returning to the global route, route planning and movement of the blindguiding robot under an arbitrary dynamic environment are achieved. Randomness of regional barrier avoidance is lowered, and the reliability of barrier avoidance under the dynamic environment is improved; generation of regional minimum values is avoided by means of a random path graph method, and the planning efficiency is taken into account without increasing the calculation amount. The blind guiding robot can walk along the global route as much as possible, and therefore the safety of the blind during movement is greatly ensured.

The invention discloses an optimization method of control parameters of a servo system of a numerical controlled machine tool. The optimization of the control parameters of the servo system of the numerical controlled machine tool affects important indicators of the machine tool such as highest moving speed, positional accuracy and repeated positioning accuracy, and further decides outline accuracy and surface quality of machining workpieces. However, the control parameters of the servo system of the numerical controlled machine tool are various, coupling between the parameters is strong, the parameters are of nonlinearity characteristic, and the parameters are complex as the number of linkage shafts of the numerical controlled machine tool increases. The invention provides an automatic optimization method of the control parameters of the servo system of the numerical controlled machine tool. The automatic optimization method of the control parameters of the servo system of the numerical controlled machine tool is capable of optimizing the control parameters of a multi-shaft and multi-servo system synchronously in real time. Meanwhile, the automatic optimization method of the control parameters of the servo system of the numerical controlled machine tool has the advantages of being high in optimum efficiency, fast in speed of convergence of the control parameters, capable of being transplanted into different numerical controlled systems to be used, and the like. The automatic optimization method of the control parameters of the servo system of the numerical controlled machine tool is capable of seeking an optimal control parameter value of the servo system of the numerical controlled machine tool.

The invention relates to the technical field of cloud computing (IaaS), in particular to a virtual machine allocation method based on particle swarm optimization. The method includes the steps of obtaining the virtual machine request and physical host resource of a data center, and constructing a virtual machine list and a physical host list; initializing a particle swarm, and setting the parameters of the particle swarm optimization; calculating the fitness value of each particle in the particle swarm, and recording the historical optimal particles of individuals and the optimal particles of population according to the particle fitness value; updating the speed and location of each particle according to an update strategy; judging whether or not the maximum time number of iterations is satisfied, if yes, outputting global optimal particle code, and if no, continuing to iterate; decoding the global optimal particle code into a virtual machine allocation scheme, and outputting the scheme. The method can improve the resource utilization when reducing the response time, and meanwhile achieve a better balance between the load balance degree and energy consumption.

The invention discloses a mobile robot path planning method based on a whale optimization algorithm. The method comprises the following steps: step S1, initializing the whale optimization algorithm, setting parameters of the algorithm, using a fitness function to obtain fitness values of a whale at all positions, and determining an initial individual optimal position and a global optimal positionof a whale population; step S2, using a new convergence factor, re-calculating a coefficient vector and updating the new position of the whale individual; step S3, calculating the fitness value of thewhale individual at the new position and comparing the fitness value with the fitness value of the original position; step S4, if the fitness value of the new position is superior than the fitness value of the original position, updating the individual best position of the whale population, and updating the global optimal position; and step S5, after reaching the number of iterations, selecting awhale path with the minimum fitness value as the optimal path for the mobile robot path planning, otherwise, executing steps S2 to S4. The mobile robot path planning method based on the whale optimization algorithm provided by the invention has high convergence precision, fast convergence speed, and can avoid falling into local optimum in the later stage of the algorithm iteration.

The invention relates to a wind power variable-pitch multi-variable fuzzy neural network PID control method. The control method includes the following steps that a fuzzy parameter setting module is used for presetting the weight of a PID neural network module; the error between a rotating speed reference value and the actual rotating speed output of a wind driven generator is calculated through a PID calculation module to obtain a reference output quantity of torque of the wind driven generator; the error between the power output value and the power reference value of the wind driven generator and the error change rate are set through the fuzzy parameter setting module to obtain a presetting parameter of the weight of the PID neural network module; through a negative gradient algorithm with a momentum factor, the weight of the PID neural network module is trained, and the reference value output of torque and the reference value output of the pitch angle of the wind driven generator are adjusted. The output power of the wind driven generator can be stabilized nearby a rated valve, and safety of a fan is ensured.

The invention discloses a milling tool path optimization method for low carbon. The milling tool path optimization method comprises the steps of firstly, building solving process of milling tool pathoptimization problem considering carbon emission by taking flat cavity processing as a research object; secondly, building processing time, carbon emission and a processing cost function of cavity milling process taking a tool path as a variable; and finally, building a multi-target tool path optimization model taking shortest processing time, minimum carbon emission and lowest processing as optimization targets, and optimizing the multi-target tool path optimization model by an improved genetic algorithm-based optimization solution method. The method comprehensively considers the targets of the processing time, the carbon emission and the processing cost, the actual application demand of an enterprise is satisfied, the tool path optimization solving process is simple and effective and iseasy to implement, the population convergence speed is rapid during path optimization solving by the proposed improved genetic algorithm, and the algorithm can be effectively prevented from being sunken to local optimal solution.

The invention discloses a calibrating method for an electronic compass. The calibrating method is used for promoting measuring accuracy of the electronic compass based on a self-adaption differential evolution algorithm and a Fourier neural network theory and is especially suitable for an orienting system with low cost and higher precision. The calibrating method comprises the following steps: utilizing a Fourier neural network to perform error modeling on the electronic compass and utilizing an improved self-adaption differential evolution algorithm to optimize a weight value of the Fourier neural network, so as to acquire an accurate error model to compensate a measuring value of the electronic compass. The error model which is established according to the calibrating method is capable of realizing accurate mapping of a sample space and has a higher nonlinear approaching capability. According to the calibrating method, a minimum local part is avoided, the defects of over-slow convergence rate and oscillation of the neural network are overcome and the influence of an outside magnetic field on the electronic compass is efficiently compensated, thereby greatly promoting the measuring accuracy of the electronic compass.

The invention discloses a live pig behavior classification method based on a BP neural network, and the method comprises the steps: collecting live pig acceleration, angular speed and attitude angle information in real time as input; obtaining a classification result according to a pre-built BP neural network model; carrying out the matching of four behavior manners of live pigs through video segment information: standing, walking, groveling and lying; jointly obtaining 6000 groups of data, and carrying out the Z-score normalization processing; selecting an LM training method for the training of a discrimination model. The method considers the attitude angle as the input variable of the BP neural network, is high in network convergence rate, and meets the requirements of instantaneity. Moreover, a local flat region can be effectively surpassed in a training process, and an expected error level is reached. The model classification precision is high. A verification result indicates that the live pig behavior discrimination model considering the attitude angle building is in highly linear relation with the actual behaviors, and the correlation coefficient is 0.992. The overall discrimination accuracy is 92.64%, and the accuracy of the discrimination model built under the condition that only the acceleration and angular speed data is considered is 86.38%, which indicates that the live pig behavior discrimination model based on the attitude angle building can provide data support for the discrimination of the health condition of the live pigs.

The invention discloses a face recognition method based on a particle swarm optimization BP network. The method includes that an image is preprocessed to eliminate external disturbance; information of the preprocessed image is projected to a feature space by means of mapping transformation and by selecting different feature extraction modes; in the training or recognition process of neural networks, each feature corresponds to one input node of each neural network, output nodes are equal to classes in number, and one output node corresponds to one class. Therefore, a fully-connected BP network is designed, wherein the number of neurons in an input layer corresponds to the number of the features of the image, the number of neurons in an output layer is the number of swarm classes, the number of neurons in a hidden layer is set as the following formal, network weight is initialized as a random value between 0 and 1, and each particle corresponds to one neuron network. According to adaptive values of the particles and variable quantities of the adaptive values, inertia weight of each particle is regulated in real time, a global optimal solution can be rapidly found out, and efficiency and accuracy of face recognition are improved finally.

The invention discloses a title push method, a storage medium, and an application system. The title push method comprises the following steps of (S1) setting at least one first database and at least one second database; (S2) receiving a title push request initiated by a user, wherein the title push request includes user ID information and at least one title practice range; (S3) obtaining a user capability index corresponding to the user ID information and the title practice range in the first database according to the user ID information and the title practice range; and (S4) determining a question group whose difficulty corresponds to the user capability index in the second database according to the obtained user capability index, and pushing the question group to the user. In the invention, customized title push is realized and the method can adapt to the dynamic process of learning capability improving.

The invention relates to a flexible job-shop scheduling optimization method, which applies the Metropolis criterion and the sinusoidal adaptive step length to a firefly algorithm so as to optimize andsolve a discrete problem. On the basis of building a mathematical model, an initial solution population of a discrete combination problem is randomly generated, then local search in an individual domain is performed according to the Metropolis criterion in simulated annealing to generate a new individual, the internal energy difference between the new individual and the original individual is calculated, the new individual is accepted according to a certain probability, and global search is performed on each generation by using the discrete firefly algorithm with the sinusoidal adaptive steplength until an optimal solution is searched. The method can better search an optimal solution of the FJSP (Flexible Job-Shop Scheduling Problem) in the global space and has better search precision, search efficiency and stability, thereby having important significance and significant engineering practical application values for solving discrete problems such as job-shop scheduling.

The invention discloses a particle swarm algorithm based photovoltaic cell panel maximum-power tracking method. The method includes: firstly, setting initial power values to determine initial positions of particle positions and the number of particles; then taking the power values corresponding to the initial positions of the particles as the optimal particle values corresponding to the particles; finally, selecting out the maximum values as optimal swarm values of particle swarms by comparison of the optimal particle values and outputting the optimal swarm values. Output voltage of a photovoltaic cell plate can be acquired according to the particle swarm algorithm, duty ratio of PWM (pulse-width modulation) is taken as updating speed of the particles, and the output voltage of the cell plate is taken as objective functions used for judging the particle positions; the updating speed of the particles is taken as output to perform PWM on a switching tube of a Boost circuit to acquire the updated particle positions, and directions are given for updating of the particles with the selected optimal values; the optimal values of the particles are searched, and an MPPT (maximum power point tracking) objective is realized. The tracking method is high in intelligent degree and tracking precision, and the cell plate capable of tracking the maximum power value points without falling into locality is the optimal.

The invention discloses a method for registering a target contour point cloud based on a monocular depth sensor and a mechanical arm. The monocular depth sensor (2) is loaded on the end of a robot arm(1). A host computer (3) controls the robot arm to accurately move a sampling position, controls the monocular depth camera to capture the target contour point cloud and preprocesses the target contour point cloud, and marks the current point cloud as a source point cloud S. The rotation angle values of each axis of the current robot arm (1) are obtained through a control cabinet (4). The robot arm (1) is modeled according to the kinematics theory, and the base coordinate system-based posture of the monocular depth sensor at the current sampling point is calculated. An improved iterative nearest point algorithm is used to complete precise registration of the point cloud in the S and D views, and the registered point cloud is marked as the source point cloud S. Steps S4 and S5 are repeatedly executed, and the point cloud under the next view is again registered until the model of the target contour point cloud is complete, and the registration then terminates.

The invention discloses an image segmentation method based on improved whale optimal fuzzy clustering, so as to mainly solve the problems of serious loss and long segmentation time after image information segmentation in the prior art. The method comprises the realization steps: 1, an image is inputted and the gray levels of all pixel points are acquired; 2, c clustering centers are selected to segment the image into c classes; 3, n whales are generated, wherein each whale has a c-dimensional vector, which represents a possible solution of a set of clustering centers; 4, with the reciprocal ofa fast fuzzy C-means clustering objective function as a fitness value, the optimal clustering center is searched; and 5, according to the searched set of clustering centers corresponding to the maximum fitness value, image segmentation is realized, pixel points with the gray levels in the same membership grade interval are classified as one class, and an image after segmentation is outputted. Through combining the optimization result and the fuzzy clustering image segmentation, the image segmentation effects are improved, and the method can be used for target detection, video monitoring and medical imaging.

The invention relates to a static reactive power compensation device based on an energy method and a control method thereof, belonging to the technical field of power transmission and distribution. The invention comprises a mutual inductor group, a filtering and signal conditioning module, an A/D conversion module, a computing module, a memory module, a phase-locked loop circuit module, a communication module and a display and keyboard operation module. The invention can evaluate and analyze possibly potential dangers of a power grid by adopting energy function method, predict the influence on the stable operation of the power grid possibly resulted from faults by establishing an anticipated accident list, carry out scientific, effective and real-time analysis and evaluation on the power grid state and carrying out integral cognition on the power grid performance as complete as possible, thereby quantitatively evaluating the operation state level of the power grid; and the invention effectively avoids an algorithm from falling to local optimum by adopting an improved particle intelligent swarm algorithm based on a fuzzy control theory.

The invention discloses an image color correction method based on a simulated annealing optimization algorithm, and belongs to the technical field of image processing and computer vision. The image color correction method includes the steps: 1) measuring RGB (red, green and blue) stimulus values of a color sample to obtain a standard value of the color sample under a standard illuminant; 2) building a color correction model and computing a sample color theoretical value; 3) adjusting the brightness of the sample color standard value; 4) computing a color difference average value serving as a target function between a converted sample color XYZ value and the sample color standard value; 5) solving a corresponding correction matrix M when the target function obtains a globally optimal solution; 6) judging whether the computed theoretical value and a sample color measuring value meet brightness constraint conditions or not, and adjusting brightness adjustment coefficient lambda according to a fixed step length and re-computing the matrix M until the brightness constraint conditions are met if the brightness constraint conditions are not met. The image color correction method has the advantages of high correction accuracy, high noise resistance and the like, and the method can adaptively adjust the brightness level before and after image correction.

The invention discloses a fault diagnosis method for a PWM inverter of a motor drive system. The fault diagnosis method is characterized in that an inverter fault diagnosis model based on both wavelet packet decomposition and an RBF neural network is designed, wavelet packet transformation is utilized to extract a feature vector of a fault signal of the inverter, and the feature vector is taken as the input quantity of the RBF neutral network; a wolf pack-simulated annealing algorithm is adopted to optimize structural parameters of the RBF neutral network; and 22 groups of learning samples and 6 groups of test samples are utilized to train and examine the RBF neutral network. Simulation experiment analysis shows that when the fault diagnosis method is used for an open-circuit fault of the PWM inverter of a three-phase motor drive system, the fault position of an inverter TGBT power tube can be accurately positioned, fault diagnosis is quick, accurate and efficient, and the fault diagnosis method contributes to improving the running reliability of the motor drive system.

The invention relates to the field of communication, in particular to an illumination communication dynamic routing ant colony algorithm based on a new probability transfer function. The algorithm is applied to a network topological graph, searches a path possibly existing between every pair of nodes in a network periodically, collects the attribute values of all routing targets on each path and records the attribute values in pheromones. The illumination communication dynamic routing ant colony algorithm based on the new probability transfer function has the advantages that the new state transfer probability function is adopted, and thus the large defect that the ant algorithm is caught into a local optimum path trap easily in path optimization is avoided; the probability transfer function is adjusted by using information weight factors in normal distribution, and thus the randomness and the blindness of the state transfer rules of the ant algorithm are reduced; the intensity of the pheromones is set by segmenting the global pheromone algorithm, and the speed of concentration increase of the pheromones on the paths where ants are concentrated excessively is relieved by introducing information amount operators based on even distribution; route routing time of carrier communication controlled by straight lamps is optimized.