Unlock instant, AI-driven research and patent intelligence for your innovation.

A mobile terminal information collection system

A technology of information collection and mobile terminal, applied in the field of length, it can solve the problems of easy to fall into local minimum point and slow convergence speed.

Active Publication Date: 2022-06-07
BEIJING UNIV OF TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problem of slow convergence speed and easy to fall into local minimum point in the optimization process of genetic algorithm, a new population selection mechanism is proposed

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A mobile terminal information collection system
  • A mobile terminal information collection system
  • A mobile terminal information collection system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0069] 12 populations are used in the project. If generalization is performed, the idea is to classify all the populations, including four categories, namely the best (Top), good (best), general (normal) and For the poor population (worse), select all the good populations and select the best, normal and worst populations in proportion. After the selection, the insufficient part is randomly generated, and the number of populations in each round remains unchanged.

[0070] The overlap range can be 0, the time window cannot be 0, the number of neurons cannot be 0, and the length of the experimental data is 89. Therefore, the overlap range can be 0-88, and the time window can be 1-89, that is, take one at a time. Data, two data, up to 89 data (that is, all data) can be obtained at one time, the minimum number of hidden layer neurons is 6, that is, 6 categories, there is no upper limit, but based on practical considerations, too many neurons The number will not bring more help to t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to an upper limb gesture recognition algorithm based on a genetic algorithm for hyperparameter optimization. The mobile terminal collects data on upper limb gestures and movements. By analyzing the collected data, six types of movement information are identified. The upper limb gesture data is a time series, and the length of the time window, data overlap rate and the number of hidden layer neurons will affect Therefore, the genetic algorithm is used to optimize the hyperparameters to find the most suitable set of solutions. The traditional genetic algorithm has a slow optimization speed and is easy to fall into a local solution. The present invention sorts the population according to the fitness function, divides the sorted population into 4 parts, namely Top, best, normal and worse, and selects all good populations And the best, normal and worst populations are selected in proportion. After the selection, the insufficient parts are randomly generated to ensure that the number of populations in each round remains unchanged. The invention speeds up the speed of parameter optimization and finds the global optimal solution at the same time.

Description

technical field [0001] The present invention proposes an algorithm for hyperparameter optimization of the collected upper limb movement posture data based on genetic algorithm, and at the same time develops a mobile terminal portable information collection system for upper limb movement posture collection. The mobile terminal information collection includes: The setting of the parameters of the attitude sensor, the real-time display of the data, the rotation of the 3D model, the saving of the data and the real-time recognition of the attitude. Three hyperparameters are simultaneously optimized by genetic algorithm. The optimized parameters are the length of the time series time window, the repetition rate of each acquired data, and the number of hidden layer neurons. At the same time, a new population selection is proposed. mechanism. This patent involves mobile terminal development and genetic algorithm optimization. Background technique [0002] At present, there are thr...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/12
CPCG06N3/126Y02D30/70
Inventor 张俊杰孙光民张子昊付晓辉姜明
Owner BEIJING UNIV OF TECH