CN111203875A

CN111203875A - Mechanical arm collision safety level detection system

Info

Publication number: CN111203875A
Application number: CN202010019824.9A
Authority: CN
Inventors: 魏大鹏; 陈羿甫; 姜星池; 杨俊�; 谢义; 周凯; 李晓霞; 洪鑫; 唐新悦
Original assignee: Chongqing University of Post and Telecommunications
Current assignee: Chongqing University of Post and Telecommunications
Priority date: 2020-01-07
Filing date: 2020-01-07
Publication date: 2020-05-29
Anticipated expiration: 2040-01-07
Also published as: CN111203875B

Abstract

The invention relates to a mechanical arm collision safety level detection system, which belongs to the field of human-computer interaction safety and comprises a sensor module, a pressure sensor and a control module, wherein the sensor module comprises a plurality of array circuit-connected sensing units of mixed fillers, and the sensing units convert pressure values into voltage signals by means of resistance changes; the data acquisition module comprises a power supply module, a microcontroller MCU module, an operational amplifier module and an output module; data are collected through a single chip microcomputer program and a data receiving program, and the single chip microcomputer program controls an I/O port to output high and low level changes for gating; the data processing and decision module comprises a K nearest neighbor KNN algorithm for identifying hardness and a probability neural network PNN algorithm for deciding a safety level; the invention can detect the collision force with large range and high precision and the collision area with wide range; and secondly, the hardness is identified by using a KNN algorithm, and the PNN algorithm decides the safety level composition, so that the reliability is greatly improved.

Description

Mechanical arm collision safety level detection system

Technical Field

The invention belongs to the field of human-computer interaction safety, and relates to a mechanical arm collision safety level detection system.

Background

A robotic manipulator sharing a workspace with humans should be able to quickly detect collisions and react safely to limit injuries caused by physical contact. Without external perception, the relative motion between the robot and the person is unpredictable and an unexpected collision may occur anywhere on the robot arm.

In the coming years, the implementation and use of the 5G standard can really realize the interconnection of everything, namely the Internet of things. At the moment, the number of robots and application scenes are increased at a very high speed, and a man-machine hybrid system is gradually and widely applied to small and medium-sized enterprises, so that the machining and assembling efficiency of a workshop is greatly improved, but the surrounding complex environment brings new challenges to the safety of the robots, and therefore sufficient safety must be ensured at the beginning of design.

Disclosure of Invention

In view of this, the present invention provides a system for detecting a collision safety level of a robot arm in a human-computer interaction process, which solves the problem that classification ambiguity and precision speed of existing safety level classification cannot be considered in the human-computer interaction process. In human-computer interaction, when the impact area of a mechanical arm in collision is too small, the impact degree of the impact force on safety level decision is small or the impact force is too large, the impact degree of the impact area on the safety level decision is small, and serious consequences are easily caused by pricking or large-volume collision. When the machine comes into contact with an object or a person, the speed is high, and data acquisition and decision making are required to be fast.

In order to achieve the purpose, the invention provides the following technical scheme:

a mechanical arm collision safety grade detection system comprises

The sensor module comprises a plurality of sensing units which are connected with the array circuit and are mixed with fillers, and the sensing units convert pressure values into voltage signals by means of resistance changes;

the data acquisition module comprises a power supply module, a microcontroller MCU module, an operational amplifier module and an output module; data are collected through a single chip microcomputer program and a data receiving program, and the single chip microcomputer program controls an I/O port to output high and low level changes for gating;

and the data processing and decision module comprises a K Nearest Neighbor (KNN) algorithm for identifying hardness and a Probability Neural Network (PNN) algorithm for deciding the safety level.

Further, the specific method for identifying the hardness through the K nearest neighbor KNN algorithm comprises the following steps: using repeated stretching experimental equipment to simulate impact to obtain the gradient characteristic G (F) of pressure changing along with time and the equilibrium steady-state pressure characteristic F_SThen, calculating the Euclidean distance from the test set point to the training set point, and judging the classification with the closest hardness;

the specific method for deciding the security level through the probabilistic neural network PNN algorithm comprises the following steps: in practice, a mechanical arm impact test is carried out, pressure size/pressure area obtained through impact and identified hardness and softness serve as input, a safety level decision based on PNN is constructed, and net (P, T, spread) is tested, wherein P and T are an input vector and a target vector respectively, and spread is the diffusion speed of PNN.

Furthermore, the collision part of the mechanical arm is provided with an elastic buffer layer, and the elastic buffer layer is made of special synthetic rubber butyl rubber and has the thickness of 2 mm.

Further, the filler is carbon nano tube and C₆₀And graphene nanoplatelets, wherein the proportion of the carbon nanotubes is not less than 23 percent, and C₆₀Not less than 64 percent, and not less than 13 percent of graphene nanoplatelets.

Further, the single chip microcomputer program controls the electric potential of 16 rows of the distributed pressure sensor by outputting high and low level changes through 4I/O ports, the gated row is set to be at a high electric potential, namely 3.3V, and the ungated row is set to be at a zero electric potential, and the command for controlling gating of each row is as follows:

further, the data receiving program comprises serial port setting, an operation command and pressure distribution imaging; the serial port is used for setting a serial port number, a baud rate, a check bit, a data bit and a stop bit of the serial port; the operation command comprises opening and closing the serial port, starting and stopping reading and storing the data of the current pressure distribution state; the pressure distribution imaging is used for rasterizing the received data and drawing the rasterized data into a pressure distribution diagram.

The invention has the beneficial effects that: the invention can detect the collision force with large range and high precision and the collision area with wide range; and secondly, the hardness is identified by using a KNN algorithm, and the PNN algorithm decides the safety level composition, so that the reliability is greatly improved.

Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.

Drawings

For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:

FIG. 1 is a flow chart of data collection according to an embodiment of the present invention;

FIG. 2 is a graph of a pressure sensor signature profile according to an embodiment of the present invention;

fig. 3 is a simulation test chart of security level decision according to an embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.

Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.

The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.

The invention provides a system for detecting the collision safety level of a mechanical arm in a human-computer interaction process, which comprises the following steps:

the sensor module is provided with a plurality of sensing units of mixed fillers 1 connected by an array circuit, and the sensing units convert pressure values into voltage signals by means of resistance changes;

as shown in fig. 1, the data acquisition module is composed of data acquisition hardware and data acquisition software, and the data acquisition hardware is composed of a power module, a specific model of which is a2356G-3FG4, a Microcontroller (MCU) module, a specific model of which is an STM32F125TY2 chip, an operational amplifier module, a specific model of which is LM435Q, and an output module, a specific model of which is DH 451; the data acquisition software consists of a single chip microcomputer program and a data receiving program, the single chip microcomputer program controls the I/O port to output high and low level changes for gating, and the data receiving program consists of serial port setting, an operation command and pressure distribution imaging;

and the data processing and decision module consists of a KNN algorithm for identifying hardness and a Probabilistic Neural Network (PNN) algorithm for deciding safety level.

In the human-computer interaction process, the system can timely judge the safety level of collision in the working process of the mechanical arm with high precision, wide range and high speed, and provides powerful data support for subsequent accident treatment and quick action.

In the embodiment, a new flexible sensor is adopted, so that the sensitivity of the sensor is improved; the special synthetic rubber butyl rubber is used as a buffer layer when the mechanical arm is in contact with a human body, so that the severity of accidents is reduced; and the flexible polyester film is adopted, so that the flexibility of the sensor is improved, and the sensor can be conveniently attached to the surfaces and joints of various complex robots. And a professional electronic device is adopted, so that the accuracy and stability of data acquisition are ensured. And a novel data processing algorithm is adopted, so that the speed and the accuracy of safety level judgment are improved.

In the present embodiment, the filler 1 is preferably a carbon nanotube or C₆₀And graphene nanoplatelets, wherein the proportion of the carbon nanotubes is not less than 23 percent, and C₆₀Not less than 64 percent, and not less than 13 percent of graphene nanoplatelets. The proportion of the mixed material is the optimal proportion, the mixed material has good mechanical stability and flexibility as a filler, when the mixed material is acted by force, the sensing unit on the flexible array circuit board is deformed, and the fillerThe density changes and the resistance changes, thereby converting the mechanical parameters into electric signals.

Preferably, the single chip microcomputer program controls the electric potential of 16 rows of the distributed pressure sensor by outputting high and low level changes through 4I/O ports, the gated row is set to a high electric potential, that is, 3.3V, the ungated row is set to a zero electric potential, and the command for controlling gating of each row is shown in table 1.

TABLE 1 control gating instruction Table

The novel data processing algorithm is characterized in that the hardness is identified by a K Nearest Neighbor (KNN) classification algorithm, a Probabilistic Neural Network (PNN) makes a decision on the safety level, and the specific method comprises the steps of firstly using repeated stretching experimental equipment to perform simulated impact to obtain a gradient characteristic G (F) of pressure changing along with time and a balanced steady-state pressure characteristic F_SThen, calculating the Euclidean distance from the test set point to the training set point, and judging the classification with the closest hardness, namely identifying the hardness by using a K nearest classification algorithm, as shown in FIG. 2; and then, carrying out mechanical arm impact test in practice, taking the pressure/pressure area obtained by impact and the identified hardness as input, and outputting safety grade classification.

As shown in fig. 3, the pressure magnitude/pressure area obtained by impact and the identified hardness and softness are used as input, and a security level decision based on PNN is constructed by using a MATLAB experimental simulation platform, wherein P and T are an input vector and a target vector respectively, and net is the diffusion velocity of PNN.

Preferably, the data receiving program includes serial port setting, an operation command, and pressure distribution imaging; the serial port is used for setting a serial port number, a baud rate, a check bit, a data bit and a stop bit of the serial port; the operation command comprises opening and closing the serial port, starting and stopping reading and storing the data of the current pressure distribution state; the pressure distribution imaging is used for rasterizing the received data and drawing the rasterized data into a pressure distribution diagram.

Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.

Claims

1. a mechanical arm collision safety level detection system, is characterized in that: comprising:

A sensor module, comprising a plurality of sensing units of mixed fillers connected by array circuits, the sensing units convert pressure values into voltage signals by means of resistance changes;

The data acquisition module includes a power supply module, a microcontroller MCU module, an operational amplifier module and an output module; data is collected through a single-chip microcomputer program and a data receiving program, and the single-chip microcomputer program controls the output high and low level changes of the I/O port to perform gating;

The data processing and decision-making module includes the identification of softness and hardness through the K nearest neighbor KNN algorithm, and the decision of the security level through the probabilistic neural network PNN algorithm.

2. The mechanical arm collision safety level detection system according to claim 1, is characterized in that: the described concrete method of identifying softness and hardness by K nearest neighbor KNN algorithm is: using repeated stretching experimental equipment to simulate impact, obtain pressure Gradient feature G(F) that changes with time, balance steady-state pressure feature F _S , then calculate the Euclidean distance from the test set point to the training set point, and judge the classification with the closest softness and hardness;

The specific method for determining the safety level through the probabilistic neural network PNN algorithm is as follows: in practice, the impact test of the manipulator is carried out, and the pressure size/pressure area obtained by the impact and the identified hardness are used as input, and a PNN-based safety level decision is constructed. , net=newpnn(P, T, spread) for testing, where P and T are the input vector and target vector, respectively, and spread is the diffusion speed of the PNN.

3. The collision safety level detection system of a robotic arm according to claim 1, wherein the collision part of the robotic arm is provided with an elastic buffer layer, and the material of the elastic buffer layer is special synthetic rubber butyl rubber, and the thickness is 2mm .

4. The robotic arm collision safety level detection system according to claim 1, wherein the filler is a mixed material of carbon nanotubes, C ₆₀ , and graphene microflakes, wherein the ratio is that carbon nanotubes are not less than 23 %, C ₆₀ not less than 64%, and graphene microflakes not less than 13%.

5. The robotic arm collision safety level detection system according to claim 1, wherein the single-chip microcomputer program controls the potential of 16 rows of distributed pressure sensors through 4 I/O ports outputting high and low level changes, and strobes the Set high potential, set zero potential for unselected rows, and the commands for each row to control gating are as follows:

6. The mechanical arm collision safety level detection system according to claim 1, is characterized in that: described data receiving program comprises serial port setting, operation command and pressure distribution imaging; Bit rate, parity bit, data bit and stop bit are set; the operation commands include opening and closing the serial port, and starting and stopping reading and saving the data of the current pressure distribution state; the pressure distribution imaging is used to The data is rasterized, and the rasterized data is drawn into a pressure distribution map.