RU2671987C1 - Manipulator - Google Patents

Abstract

FIELD: robotics.SUBSTANCE: invention relates to the field of robotics, and can be used in industrial processes for the repetitive operations execution automation. Manipulator contains static and dynamic blocks, made with the possibility of connection to each other by the connecting pads with the formation of a chain of blocks of a given configuration with the dynamic block as the manipulator terminal device. Each dynamic unit is made with possibility of linear or rotary movement and location in two end positions and contains a pneumatic or hydraulic drive and the drive control module, configured to transmit control signals and supply voltage to the next block, and the static blocks are made in the form of an extension cord or elbow. At that, the static blocks connection pads contain an electrical connector enabling the control signals and supply voltage transmission to the next block, and each dynamic block is configured to set its position by the binary code, at that, each bit is responsible for the state of one block.EFFECT: invention is universal, simple in design and control, and provides automation of production processes.4 cl, 6 dwg

Description

Technical field

The invention relates to the field of robotics and can be used in production processes to automate the performance of repetitive operations.

Prior art

Known patent US 4766775 (A), published 08/30/1988, "Modular robot manipulator)). A modular robotic arm is disclosed. The manipulator uses a basic module, any number of usually the same, but suitable sizes of articulation modules, expansion modules that may be required, and the working body of the robot to perform the task assigned to the manipulator. The hinge modules are designed to have three different movements around three different axes that intersect at a common point, to implement the range of possible movement that defines the imaginary sphere. This module is also designed to provide a separate torque through the output coupling to the next module. To connect the modules, coil-switched locks are used. The working body of the robot or the grip uses a rotating shaft and a follower connected through a cable and a pulley to control the grips.

Known patent US 6084373 (A), published 04.07.2000, "Reconfigurable modular joint and robots produced therefrom)). The present invention provides a reconfigurable modular drive unit, which can be used as the basis for building and configuring robotic and automated systems as an interconnected network of individual nodes, each node representing a single modular connection. Each modular hinge can be quickly configured in a pivot, pitch or yaw configuration. A large number of different robot structures can be built using a small number of modular connections in any of these three configurations. The modules are equipped with quick-connect mechanisms, so the new robotic structure can be assembled in a few minutes. A robot or other automatic system assembled from such modular connections is a real reconfigurable and modular system. The management system is decentralized. Each modular connection is equipped with its own integrated control system and electronics. Each modular connection includes an engine and associated sensors. An integrated control system is provided, including a power amplifier for the engine, a sensor interface, a microprocessor and a communication circuit. The only external connection to each module is the communication bus between the modules, the host computer and the power bus. A reconfigurable modular connection, comprising: a first housing having a drive means, an axis of rotation and opposite ends; a first coupling mechanism for releasably connecting the coupling member to one of said opposed ends of said first housing; a second coupling mechanism for releasably connecting the coupling member to another of said opposed ends of said first housing; a removable, attachable third coupling mechanism for releasably connecting the coupling member to said first housing in a plane perpendicular to said axis of rotation.

Known patent application US 2014121803 (A1), published 01.05.2014, "Modular And Reconfigurable Manufacturing Systems". A manufacturing system comprising: several manipulation modules, each of which includes means for mechanically and electrically connecting the module to another component in the system; the working body of the robot, which includes means for mechanically and electrically connecting the working body to another component in the system, in which the modules and the working body can be mechanically and electrically connected in various ways to change the configuration of the system and production tasks that the system can perform. Manipulation modules include a linear manipulation module configured to provide linear movement within the system. The manipulation modules further include a rotation module configured to provide angular rotation in the system. Each manipulation module has upper and lower sides, and means for mechanically and electrically connecting the module are provided on both sides of the module so that each module can be mechanically connected to two other components in the system. Manipulation modules and a working body contain a microprocessor in which microprocessors are connected to a common bus for distribution of system control. Additionally contains a main controller configured to communicate with the manipulation modules and the working body, and the manipulation modules and the working body are configured so that they automatically report their respective positions and orientations in the system to the main controller.

In industrial automation, it is necessary to mechanize repetitive actions on workpieces, for example, moving from one unit to another, feeding to machines and presses, reorienting, etc. At the same time, the mechanism that performs this action does not have to be reprogrammable (the movement is always the same).

In the process of designing a production line, the choice arises: 1) use a universal solution such as a robotic arm or 2) design an individual solution. The first option is fast, but expensive, the second is time-consuming, but economical. Often, due to the need to reduce the cost of production, the first option is not suitable (a robotic arm is an expensive universal solution whose functionality is redundant for a given task). At the same time, the second option requires the continuous work of a highly qualified professional, the result of which is an individual solution.

This invention is a platform that has the necessary versatility to build private solutions without excess cost.

The technical task is to create a technical solution for the automation of production processes.

Disclosure of the invention

To solve the technical problem, a manipulator is proposed, characterized by static and dynamic blocks, made with the possibility of connection with each other, with each dynamic block containing a control module, a drive and accepting only two end positions, the configuration of the manipulator for each production task is calculated algorithmically according to the specified target positions manipulator, spatial constraints and block parameters. Each dynamic block can be set to the value of its movement. The dynamic unit drive contains a pneumatic or hydraulic drive. The position of the manipulator is set by a binary code, and each bit is responsible for the state of one block.

Description of drawings

In FIG. 1 shows a general diagram of a manipulator.

In FIG. 2 shows a dynamic tilted block of the manipulator.

In FIG. 3 shows a dynamic swivel block of a manipulator.

In FIG. 4 depicts a conditional illustration of the manipulator in two possible states, the distance to the target points and the area of restriction.

In FIG. 5 and FIG. 6 shows a manipulator configured for a production task in two possible positions.

The implementation of the invention

The manipulator contains static and dynamic blocks 1 (Fig. 1). The dynamic unit from the static one is characterized by the presence of a drive 2 and a control module 3 (Fig. 2, 3). Dynamic blocks include blocks for linear, rotational and inclined movement. All blocks 1 have an "input" and "output", which are universal connecting pads 4, allowing you to connect any block to any other block. A terminal device (not shown) is also a dynamic unit, but has only an “input”.

The drive 2 in this implementation of the invention includes a linear or rotary pneumatic actuator connected to the pneumatic distributor 7 (Fig. 2, 3). A hydraulic drive can also be used as drive 2.

Universal connecting pads 4 contain an electrical connector 5, for transmitting control signals 8 and supply voltage to each unit 1, and a pneumatic quick disconnect connection 6, for transmitting air 9 under pressure for operation of the actuators 2.

Static blocks are designed to lengthen or rotate the chain of blocks and improve the configuration of the manipulator, and can be performed as an extension cord or elbow. Static blocks simultaneously transmit all signals 8, supply voltage and air 9 to the next block. To do this, in the fixed unit, the connectors at the “input” are directly connected to the connectors at the “output”.

The control module 3 is based on the microprocessor and is designed to control the drive 2 and transmit control signals to the next block 1. The control module 3 is installed in each dynamic block.

The dynamic block shown in FIG. 2 and FIG. 3, in order to simplify the design, it can take only two positions, which is realized through the use of a pneumatic drive 2. The design of the dynamic unit can be supplemented by adding a latch or regulator (not shown), with which you can set the end position of the dynamic unit. Thus, for a rotation block or an inclined block, you can set the angle of rotation and tilt, for a linear displacement block, you can set the offset distance. This leads to greater versatility of dynamic blocks in the configuration of the manipulator.

Control logic

Each dynamic block can take only two end positions. The target position of the block is set by one data bit (0 or 1). A position corresponding to zero is conventionally called closed, and the corresponding unit is open.

To set the position of the entire chain of blocks from external automation, it is enough to transmit a sequence of bits. In particular, for a chain of eight dynamic blocks, one byte of information can be transmitted. For example, a byte with a value of 0b01010101 (in the binary system) will mean the target position of the manipulator, in which the 1st, 3rd, 5th, 7th dynamic blocks should go to the open position, and the 2nd, 4th 6th, 8th - closed.

The processing algorithm of the control team. A control command containing the specified sequence of bits is sent to the root block. The root block is the first in the chain; further processing of the command is the same for all blocks. The next block that received the command: remembers the value of the first bit of the sequence as its target position; performs a binary shift of the sequence to the right by one bit; sends the command modified in this way to its “output” - to the next block, or to the terminal device.

Thus, the command is transmitted via the internal bus from block to block and reaches the last block or terminal device. Moreover, the command length for the terminal device can be arbitrary (not limited to one bit).

Protocol implementation

Fast Ethernet or Gigabit Ethernet packet technology can be used for both the internal bus device and the external interface of the units. Over the Ethernet channel, a protocol for transmitting control commands based on the UDP / IP protocol is implemented.

In the simplest case, two main commands are implemented:

1) Setting the target position. External automation sends a sequence of bits of the target position of the manipulator, the blocks respond about the result of the command.

2) Interrogation of the status of the manipulator. External automation sends a request, blocks send a response containing the current state (open / closed), as well as additional service information if necessary (error codes, diagnostic information, etc.)

Addressing.

Each unit has two independent Ethernet interfaces, one at the input and one at the output. For the purpose of unification during production, standard addresses for each interface are registered in each block. If the unit is used as the root, then the IP address is used for communication with external automation (if necessary, this address can be changed). The rest of the system implies the absence of any mandatory configuration - when connecting the units to each other, they are immediately ready to communicate with each other by sending broadcast Ethernet frames.

Due to the use of a high-speed data channel, the command transmission rate to one block will be quite small. For Fast Ethernet, it will be ~ 6 microseconds (command length ~ 640 bits / speed 100 Mbit / s), when using Gigabit Ethernet ~ 0.6 μs. The processing speed of a command with one block will be no more than 1 μs when using modern microprocessors. Thus, with a chain length of 10 dynamic blocks, the command will reach the last block in 70 and 16 microseconds for Fast Ethernet and Gigabit Ethernet, respectively.

Implementation of the optimal configuration search algorithm

The task of searching for a manipulator configuration is reduced to the task of search engine optimization (also see discrete combinatorial optimization in the literature) and can be solved by one of the known methods, for example, using the genetic algorithm that is known.

Formulation of the problem:

There are N different types of blocks. For each type of block, two 4 × 4 matrices are known that describe the geometric transformation between the input and output of the block in each of its two final positions — M1, M2. For each type of block, a geometric collision shell is also known. Block parameters include block mass, dynamic parameters, and geometric parameters.

A 4 × 4 matrix of the base (base) is given, on which the first block is mounted - MO.

There are m target positions 10 (Fig. 4) of the manipulator flange and target matrices T1 ... Tm are defined.

Given the geometric shell of the obstacles 11 (Fig. 4), with which the manipulator should not collide when performing manipulations.

It is necessary to find the configuration that best allows you to bring the manipulator flange to all specified target positions (Fig. 4).

Decision:

It is described by the genome in the form: {d1 ... dn}, where di is the number corresponding to the type number of the block. i is the block position in the sequence, n is the maximum number of blocks.

Each type of block is assigned its own number from 0 to N.

The fitness function is programmed. The fitness of an individual is assessed by the criterion - the sum S of the minimum scalar differences 12 (Fig. 4) between the target matrices and the matrices that the manipulator flange can reach in this configuration (corresponding to the genome), without encountering itself and obstacles. The flange matrix is calculated through the direct kinematics function for the kinematic chain - by multiplying the corresponding block transformation matrices (M1, M2 depending on the state of each block).

In order to improve the quality of the search, as well as to take into account additional criteria during optimization, the fitness function can be modified as necessary, in addition, an individual selection strategy can be applied that provides a faster or better search result.

A crossover (crossover operator) and a mutator (mutation operator) are programmed according to one of the known methods.

A further solution to the problem is known, evolutionary algorithms are described in the literature: “MODERN ALGORITHMS FOR SEARCH OPTIMIZATION. Algorithms inspired by nature ”, A.P. Karpenko, MSTU N.E. Bauman.

To automate a well-known production process, first, according to the specified target positions of the manipulator, spatial restrictions and block parameters, the manipulator configuration is calculated, which allows the manipulator flange to be brought as close to the target positions as possible, then a manipulator corresponding to the calculated configuration is assembled from the available blocks. In FIG. 5 shows an assembly of two manipulator blocks. The blocks have fixed parameters, this imposes certain restrictions on the capabilities of the manipulator, namely that the manipulator flange may not reach the precisely set target position, this is compensated by the setting of the production line.

The manipulator has a simple design and control, and its configuration is calculated for each production task, which allows it to be used to automate production processes.

Claims (4)

1. Manipulator containing static and dynamic blocks, configured to connect to each other by connecting pads with the formation of a chain of blocks of a given configuration with a dynamic block as the terminal device of the manipulator, with each dynamic block being made with the possibility of linear or rotary movement and arrangement in two final positions and contains a drive and a drive control module configured to transmit control signals and supply voltage in the wake the main unit, and the static units are made in the form of an extension cord or elbow, while the connecting pads of the static units contain an electrical connector that provides the transmission of control signals and supply voltage to the next unit, and each dynamic unit is configured to set its position with a binary code, each a bit is responsible for the state of one block. 2. The manipulator according to claim 1, characterized in that each dynamic unit can be set to the value of its movement. 3. The manipulator according to claim 1, characterized in that the drive of the dynamic unit is made in the form of a pneumatic drive. 4. The manipulator according to claim 1, characterized in that the drive of the dynamic unit is made in the form of a hydraulic drive.

Images