Component for a machine or industrial plant and method for controlling a component in a machine or industrial plant

The network communication device addresses packet errors in real-time networks by adapting redundancy levels post-error detection, ensuring efficient and reliable data transfer with minimal downtime and energy use.

DE102017215100B4Active Publication Date: 2024-12-05ROBERT BOSCH GMBH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
DE102017215100
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2017-08-30
Publication Date
2024-12-05
Estimated Expiration
2037-08-30

AI Technical Summary

Technical Problem

Current data transmission methods in real-time capable networks, such as Sercos III and Profinet, fail to prevent successive data packet errors, leading to potential failures in industrial systems due to unshielded cables or radio transmission disturbances, which cannot be compensated by existing error corrections, resulting in costly downtimes.

Method used

A network communication device that tolerates packet errors in one transmission cycle, signals their presence, and adapts redundancy levels in the next cycle based on error detection, ensuring efficient error correction without repeat transmissions, allowing for very short data transmission cycles and energy efficiency.

Benefits of technology

This approach minimizes component failures by effectively handling packet errors, enabling short cycle times, reducing energy consumption, and ensuring reliable data transfer in disturbed transmission mediums, thus preventing critical multiple errors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

A component for a machine or industrial plant and a method for controlling a component in a machine or industrial plant are provided. The component comprises a network communication device for controlling the transmission of data in a data network of the machine or industrial plant, in which data packets of a predetermined packet length are cyclically transmitted between the component and at least one other component of the machine or industrial plant, without allowing retransmission of time-critical data within a transmission cycle. The network communication device tolerates any packet errors in a received data packet that was sent in the preceding transmission cycle and signals the presence or absence of a packet error to at least one other component in the following transmission cycle.wherein the network communication device determines whether or not at least one other component has signaled a packet error for a data packet sent by the component in the preceding transmission cycle, and wherein, depending on the result of the determination, the network communication device adjusts the number of redundancy data in the data packet to be sent in the following transmission cycle, which increases the probability of correcting bit errors in the time-critical data to be sent.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The present invention relates to a component for a machine or industrial plant and a method for controlling a component in a machine or industrial plant, with which successive data packet errors are avoided during data transmission between components of the industrial plant without data packet repetition, so that secure real-time capable data transmission is possible.

[0002] In an industrial plant, control devices are used, for example, to control transport systems; to control tools such as welding tools, screwdriving and / or drilling tools, riveting tools, etc.; to control sensors; to control actuators such as linear motors or rotary machines, etc. In these cases, actions or tasks, and thus the control of a component of the industrial plant to be controlled, such as a tool, are usually dependent on the actions or results of another component of the industrial plant. Therefore, real-time data transmission is often required between the control device and the components of the industrial plant to be controlled.According to the DIN 44300 (Information Processing), Part 9 (Processing Sequences) standard, which has since been replaced by DIN ISO / IEC 2382, real-time is defined as the operation of a computer system in which programs for processing incoming data are constantly operational, such that the processing results are available within a specified time period. Depending on the application, the data can be generated randomly or at predetermined times.

[0003] In real-time networks, the transmission of time-critical data packets must therefore be guaranteed within a specific time window. In many real-time systems, this time window is determined by communication cycles in which data is exchanged periodically or cyclically. The receipt of the data packets must generally not be delayed beyond this time window. In real-time networks, this ensures that valid control data and status information are available at specific times and can be processed further.

[0004] In industrial automation, an example of real-time capable networks, control and status data are continuously exchanged between a central control unit and multiple sensors or actuators. This requires real-time capability with guaranteed maximum transmission times, as well as high data reliability. Therefore, the data transmission method must ensure that the data is successfully delivered at a specific time.

[0005] However, the problem is that in many real-time capable networks, the available transmission capacity is not fully utilized by a single time-critical service. The remaining transmission capacity can be used, among other things, for transmitting additional, non-time-critical data to the same network node and / or transmitting time-critical and non-time-critical data to other network nodes using the same transmission medium and / or for transmission breaks to reduce energy consumption. The timeframe and allocation of transmission capacity are usually controlled centrally.

[0006] Data packet transmission errors cannot be completely eliminated with current transmission methods such as Sercos III, Profinet, EtherCAT, etc. Depending on the transmission medium, such as unshielded cables or wireless transmission, and possibly shortened data transmission cycle times, one or more packet errors may occur consecutively. Currently available error correction tools cannot compensate for this, so packet errors can lead to the failure of at least one component of the industrial plant or even the entire industrial plant. This results in costly downtimes of the industrial plant.

[0007] Therefore, the object of the present invention is to provide a component for a machine or industrial plant and a method for controlling a component in a machine or industrial plant, with which the aforementioned problems can be solved. In particular, a component for a machine or industrial plant and a method for controlling a component in a machine or industrial plant are to be provided, with which failures of components of the industrial plant can be minimized.

[0008] This object is achieved by a component for a machine or industrial plant according to claim 1. The controller comprises a network communication device for controlling a transmission of data in a data network of the machine or industrial plant, in which data packets with a predetermined packet length are transmitted cyclically between the component and at least one further component of the machine or industrial plant, without a retransmission of time-critical data of the data packet being possible in a transmission cycle, wherein the network communication device is designed to tolerate any packet error in a received data packet that was sent in the previous transmission cycle and to signal the presence or absence of a packet error to the at least one further component in the following transmission cycle.wherein the network communication device is configured to determine whether or not the at least one further component has signaled a packet error for a data packet sent by the component in the preceding transmission cycle, wherein the network communication device is configured to adapt, depending on the determination result, in the data packet to be sent in the following transmission cycle, the number of redundancy data which increase a correction probability of bit errors in the time-critical data to be sent.

[0009] This component ensures transmission quality by applying error correction and error detection, but not explicitly preventing all packet errors, only critical multiple errors. This approach simultaneously achieves the following advantages.

[0010] The described component enables very short data transmission cycle times, as no retransmissions are required. Such very short data transmission cycle times are, in particular, less than or equal to 1 ms.

[0011] In addition, the described component enables moderate redundancy for error correction on average over time, since the redundancy only needs to be increased in the rare cases when packet errors have occurred.

[0012] In addition, the component offers great energy efficiency, since the additional redundancy only needs to be sent and evaluated in rare cases.

[0013] In addition, the described component realizes an efficient avoidance of critical multiple errors, which are the decisive quality feature in certain applications, e.g. in industrial communication in industrial plants.

[0014] This makes the component suitable for use in future real-time capable networks with short cycle times that use a disturbed transmission medium with an increased bit error probability, such as radio transmission, unshielded cables, etc., and where retransmission in a cycle is not possible and therefore not an option for dealing with packet transmission errors.

[0015] Advantageous further embodiments of the component are specified in the dependent claims.

[0016] The network communication device may be configured, if no packet error is detected in the preceding transmission cycle, to increase the number of time-critical data in the data packet to be sent in the following transmission cycle or to include non-time-critical data which are to be transmitted but for which a time delay is not critical, and wherein the network communication device is configured, if a packet error is detected in the preceding transmission cycle, to include the redundancy data in the data packet to be sent in the following transmission cycle instead of the non-time-critical data or a part of the time-critical data.

[0017] In a specific embodiment, the network communication device is further configured to include in the data packet: first information about whether or not a data packet sent by the at least one further component to the component in the previous cycle had a packet error, and second information about the current configuration of the data packet to be sent in the current transmission cycle, wherein the second information can be used by the at least one further component for error correction of the time-critical data. The first information can be contained in one bit, and the second information can specify the number of check bits in the data packet. Additionally or alternatively, the redundancy data are check bits.

[0018] It is conceivable that the network communication device is also designed to examine the time-critical data for bit errors and, if at least one bit error is present, to carry out error correction of the time-critical data on the basis of the second information and the redundancy data.

[0019] In one embodiment, the component further comprises an application configured to use the time-critical data, wherein the network communication device is further configured to transfer the time-critical data to the application if the network communication device has not detected a packet error, and wherein the network communication device is further configured to inform the application that transmission errors have occurred and no valid data is available in this transmission cycle if the network communication device has detected a packet error.

[0020] According to a specific embodiment, the component is a control unit configured as the at least one further component for controlling at least one tool of the industrial plant. According to another specific embodiment, the component is a control unit of a machine, wherein the control unit is configured as the at least one further component for controlling at least one sensor and / or actuator, or wherein the component is a control unit of a vehicle, wherein the control unit is configured to control a steering assistance system or a brake booster system of the vehicle.

[0021] At least two of the components described above can be part of a data network which also has a wired or wireless transmission medium, wherein the at least two components are connected to one another via the transmission medium in order to be able to carry out a transmission of data in the data network, in which data packets with a predetermined packet length are transmitted cyclically between one of the components and at least one further component of the machine or industrial plant, without a retransmission of time-critical data of the data packet being possible in a transmission cycle.

[0022] In the data network, a control device may be designed to control at least one tool of the industrial plant as the further component and / or to control at least one sensor and / or actuator as a further component, wherein the control device is designed such that the control device as the central controller can exchange data packets bidirectionally with each further component, and wherein each further component is designed such that each of the further components can exchange data packets only with the central controller.

[0023] The object is further achieved by a method for controlling at least one component in a machine or industrial plant according to claim 10. In this case, a network communication device for controlling a transmission of data in a data network of the machine or industrial plant is provided for the at least one component, in which network communication device data packets with a predetermined packet length are transmitted cyclically between the component and at least one further component of the machine or industrial plant, without a retransmission of time-critical data of the data packet being possible in a transmission cycle.The method comprises the steps of: tolerating, with the network communication device, any packet error in a received data packet that was sent in the previous transmission cycle, determining, with the network communication device, whether or not the at least one further component has signaled a packet error for a data packet sent by the component in the previous transmission cycle, adapting, with the network communication device, depending on the result of the determination, for the following transmission cycle in the data packet to be sent, the number of redundancy data that increase a correction probability of bit errors in the time-critical data to be sent, and including signaling data in the data packet to be sent that signal the presence or non-presence of a packet error to the at least one further component in the following transmission cycle.

[0024] The process achieves the same advantages as previously mentioned with regard to the component.

[0025] Further possible implementations of the invention also include combinations of features or embodiments described above or below with respect to the exemplary embodiment not explicitly mentioned. In this case, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

[0026] The invention is described in more detail below with reference to the accompanying drawings and exemplary embodiments. They show: Fig. 1 shows a highly simplified view of an industrial plant in which components of the industrial plant according to the first embodiment are connected in a data network for data transmission; Fig. 2 shows a highly simplified structure of a coded data packet with weak channel coding according to the first embodiment; Fig. 3 shows a highly simplified structure of a coded data packet with strong channel coding according to the first embodiment; and Fig. 4 to Fig. 6 each shows a diagram illustrating a method for controlling a component in an industrial plant according to the first embodiment.

[0027] In the figures, identical or functionally equivalent elements are provided with the same reference numerals unless otherwise indicated.

[0028] Fig. 1 shows an industrial plant 1, which may, for example, be a production line for vehicles, furniture, buildings, etc., or a chemical plant, etc., in which media or workpieces 4, 5 can be transported and / or processed with at least one tool, such as an agitator, a welding tool, a screwing and / or drilling tool, a punching tool, a riveting tool, etc. However, the industrial plant 1 may alternatively be or comprise a machine, such as a printing press, a CNC machine (CNC = Computer Numerical Control), etc., which has at least one controlled actuator or drive for, for example, a controlled printing roller, etc. The machine is optionally a vehicle, which has a control system for a steering power assist system or a brake power booster or for other components of the vehicle. Any other examples of a machine and / or industrial plant 1 are conceivable.

[0029] In the specific example of Fig. 1, the industrial plant 1 has a control unit 10 as a component of the machine or industrial plant 1, a transmission medium 20, and at least one further component 31, 32, 33. The control unit 10 and the at least one further component 31, 32, 33 of the machine or industrial plant 1 jointly use the transmission medium 20 and form a data network. Both the control unit 10 and the at least one further component 31, 32, 33 can each bidirectionally transmit data in the form of data packets 21, 210 via the transmission medium 20 for applications 101, 311 to 31N, 321, 331 to 33N. For this purpose, the control unit 10 has a network communication device 100. The at least one further component 31, 32, 33 each comprises one of the network communication devices 310, 320, 330, as shown in Fig. 1. An interference 25 may occur in the transmission medium 20, which may cause errors in the transmission of the data packets 21, 210. In the case of an unshielded cable as the transmission medium 20 or in the case of a radio transmission via the transmission medium 20, such an interference 25 is, for example, extraneous radiation resulting from electromagnetic radiation, which emanates in particular from devices, lightning strikes, etc.

[0030] The control unit 10 has, as application 101, for example, a microcontroller, a memory device, and software, etc. The software implements, for example, a comparison device that compares actual variables during operation of the industrial plant 1 with target variables. Furthermore, an evaluation device can be implemented that, based on the result of the comparison device, produces new control data that is to be transmitted in the data packets 21, 210 in real time or non-real time via the transmission medium 20 to the at least one further component 31, 32, 33. The real-time control data is generally referred to below as time-critical data. The non-real-time control data is generally referred to below as non-time-critical data.

[0031] The first further component 31 is, for example, a device that has, as at least one application 311 to 31N, at least one element to be controlled, in particular at least one actuator or drive. The at least one actuator or drive can drive an axis in a linear movement or, alternatively, in a rotational movement. In particular, the first further component 31 can be a robot, a transport device, a lathe, a screwing tool, an agitator, etc. The control of the first further component 31 and / or its at least one application 311 to 31N is carried out at least partially on the basis of the control data of the control unit 10, which the first further component 31 receives in the data packets 21, 210.

[0032] The second further component 32 is, for example, an operating device, which is to be controlled with regard to which displays it shows, or which is controlled by inputs from, for example, a user. The operating device has, as application 321, for example, a keyboard and / or a mouse and / or a touch-sensitive or touch-insensitive screen, etc., or combinations thereof. In particular, the operating device is a control panel, a personal computer, a laptop, a smartphone, a tablet PC, etc. The control of the second further component 32 and / or its at least one application 321 is carried out at least partially on the basis of the control data of the control unit 10, which the second further component 32 receives in the data packets 21, 210.

[0033] The third additional component 33 is, for example, a detection device that has at least one sensor as an application 331 to 33N. The at least one sensor can, as needed, detect actual variables of physical or chemical properties of the workpieces 5, 6 or of dimensions or distances from tools to the workpieces 5, 6, etc. Any desired actual variables are conceivable. The control of the third additional component 33 and / or its at least one application 331 to 33N is carried out at least partially based on the control data of the control unit 10, which the third additional component 33 receives in the data packets 21, 210.

[0034] The other components 31, 32, 33 each send data packets 21 or data packets 210 back to the control unit 10 as a component of the industrial plant 1. The data packets 21, 210 contain, for example, results of the control processes that have occurred in the other components 31, 32, 33 based on the control data of the control unit 10. Each data packet 21, 210 corresponds to one of at least two channel codings of the transmission medium 20. The at least two channel codings each result in the same coded packet length, as in Fig. 2 and Fig. 3 shown.

[0035] Fig. Figure 2 shows the structure of a data packet 21 in more detail. The data packet 21 has four different sections, which are arranged chronologically in a predetermined packet length N. The predetermined packet length N corresponds to a number of N bits. The data packet 21 has a first section for signaling data 211, a second section for time-critical data 212, a third section for non-time-critical data 213, and a fourth section for first redundancy data 214. The first and second sections together have a number L1 of bits. The non-time-critical data 213 has a number L2 of bits. Thus, the first redundancy data 214 has a number N - L1 - L2 of bits. The signaling data 211 contains information 211A, 211B, whether the previous transmission was successful or not, and information indicating the current configuration of the error correction, for example, the number of check bits, as described in relation to Fig. 4 to Fig. 6 is described and illustrated in more detail.

[0036] According to Fig. 3, however, a data packet 210 only has three different sections, which are also arranged chronologically in the predetermined packet length N. The predetermined packet length N corresponds to the number of N bits. The data packet 210 also has the first section for signaling data 211 and the second section for time-critical data 212. However, the data packet 210 has a third section for second redundancy data 215. Here, the first and second sections together again have the number L1 of bits. Thus, the second redundancy data 215 has a number N - L1 of bits.

[0037] Data package 21 according to Fig. 2 corresponds to a weaker channel coding than the data packet 210 according to Fig. 3. The channel coding parameters are selected such that the number of useful bits for the data packet 210 with the stronger channel coding corresponds at least to the length of the time-critical data 212. The length of the time-critical data 212 results from the number of bits for the time-critical data 212.

[0038] Normally, moderate channel coding with a relatively small number of redundant bits is used in the operation of the network communication devices 100, 310, 320, 330. This low redundancy makes it possible to transmit L2 bits for non-time-critical data 213 in addition to the L1 bits for the signaling data 211 and the time-critical data 212. Thus, a data packet 21 is normally sent.

[0039] The moderate channel coding according to data packet 21 is selected such that, given the expected bit error probability of transmission over transmission medium 20, all transmission errors can be eliminated in most cases. However, the packet error probability PER1 does not necessarily correspond to the desired error probability of system-critical multiple errors.

[0040] The stronger channel coding according to data packet 21 is selected such that, despite the higher number of redundant bits, the same coded predetermined packet length N is present as with the moderate or weaker channel coding. Therefore, the number of useful bits is reduced. This can be achieved, on the one hand, by temporarily suspending the transmission of non-time-critical data 213. Alternatively or additionally, the transmission of time-critical data 212, which can be dispensed with at the present time, can be suspended in the hope that it will be transmitted correctly again in the next cycle. The parameters of the stronger channel coding are selected such that the error probability after decoding is lower than the desired probability of critical multiple errors.

[0041] If weak channel coding is normally used, i.e., data packet 21 is used, the probability for a single packet error is PER1. If such a single error is detected by control unit 10 or one of components 31, 32, 33, the stronger channel coding is applied in the next cycle, i.e., data packet 210 is used. The conditional probability that a first error occurs and a second error occurs is PER2. The probability for a double error is therefore PER1·PER2.

[0042] The following is based on Fig. 4 to Fig. 6 describes the communication that takes place during operation of the industrial plant 1 between the control unit 10 as a component of the industrial plant 1 and the other components 31, 32, 33. This implements a method for controlling a component 10, 31, 32, 33 in the machine or industrial plant 1. The method can be integrated into any suitable transmission method, such as Sercos III, Profinet, EtherCAT, etc.

[0043] In Fig. 4 to Fig. 6, the time axis runs from bottom to top. A transmission cycle K has a duration T_K. A transmission cycle K+1 following the transmission cycle K has a duration T_K+1. The transmission capacity provided for the periodic or cyclic communication between the control unit 10 and the further component 31, as an example in the transmission cycles K and K+1, is shown in Fig. 4 by the rectangles for the packages 21A. In Fig. 5 and Fig. 6, packets 21B, 210A are also sent, as explained below.

[0044] In the method, the data to be transmitted, i.e., time-critical and non-time-critical data 212, 213, are transferred from one of the applications 101, 311 to 311N to a MAC layer 1001, 3101 (MAC = Media Access Control) in the respective network communication device 100, 310. The MAC layer 1001, 3101 controls how the data packets 21A, 21B, 210A are assembled depending on the system state. The system state depends, for example, on which tasks are to be performed in the machine or industrial plant 1.

[0045] The respective data packets 21A, 21B, 210A are then passed to a channel coding / decoding layer 1002, 3102, where, depending on the setting, weak or strong redundancy is added. This results in the data packets 21A, 21B, 210A in Fig. 2 and Fig. 3 shows different types of channel coding 200.

[0046] As a result, a data packet 21A contains time-critical and non-time-critical data 212, 213 as well as redundancy data 214 and, in the signaling data 211, a signaling 211A that the data of a data packet of the previous transmission cycle was error-free.

[0047] A data packet 21B is constructed in the same way as a data packet 21A, except that in the signaling data 211, a signaling 211B shows that the data of a data packet of the previous transmission cycle was not error-free and thus a packet error 26 ( Fig. 5 and Fig. 6) was present.

[0048] A data packet 210A contains time-critical data 212 and redundancy data 214 and, in the signaling data 211, a signaling 211A that the data of a data packet of the previous transmission cycle was error-free.

[0049] The encoded data packets 21A, 21B, 210A are then transmitted on the transmission medium 20, whereby only the reserved transmission resource, in particular a specific time slot, can be used. If the disturbance 25 occurs on the transmission medium 20, bit errors can occur in the data packets 21A, 21B, 210A during transmission over the disturbed medium 20.

[0050] The respective data packet 21A, 21B, 210A is received and decoded at the network node or control unit 10 or one of the components 31, 32, 33. In most cases, any bit errors can be eliminated through error correction. If this is not possible, transmission errors should at least be reliably detected. Well-known methods such as Cyclic Redundancy Check (CRC) are suitable for this purpose.

[0051] However, there are also options for combined error correction and error detection. The information regarding whether the transmission was successful is communicated to the MAC layer 1001, 3101 of the network node or the control unit 10 or one of the components 31, 32, 33. The MAC layer 1001, 3101 then forwards the received payload data, namely the time-critical and / or non-time-critical data 212, 213, to the application layer 1001, 3101 or discards erroneous data. Furthermore, the network node or the control unit 10 or one of the components 31, 32, 33 sends the success or failure of the channel decoding via layer 1002, 3102 along with its payload in the reverse direction. The basic procedure, consisting of assigning the packet structure, channel coding, transmission with the assigned resource, and decoding at the receiver, is identical to the forward direction.

[0052] The bidirectional communication in the transmission medium 20 can, as in Fig. 4 to Fig. 6, can be implemented as a half-duplex method, in which only one direction can be transmitted at a time. Alternatively, the bidirectional communication in the transmission medium 20 can be implemented as a full-duplex method, in which both directions can be transmitted simultaneously. The remaining transmission capacity resulting from the free time slots between the reserved slots could be used for additional transmission services or other network nodes.

[0053] In Fig. Figure 4 shows a transmission without packet errors for transmission cycles K and K+1. Therefore, only the weaker channel coding is used in both transmission directions. Consequently, data packets 21A are sent in each case. Successful error correction is communicated to the MAC layer 1002, 3102 as message 216A and to the sender via ACK signaling 211A. As long as the control unit 10 or component 31 receives an ACK as signaling 211A, the weaker channel coding is used for the following cycle.

[0054] In contrast, the example of Fig. 5, a packet error 26 occurs in the forward direction in transmission cycle K. If such a packet error 26 cannot be corrected but can be detected in cycle K during channel decoding in the network node, such as component 31, this is communicated to the MAC layer 1002, 3102 as message 216B. The MAC layer 3102 then discards the received payload data 212, 213. The packet error 26 is communicated to the application layer, which then reacts accordingly. For the non-time-critical data 213, a retransmission can be requested if necessary. It is assumed that, despite the packet error 26, the ACK signaling 211A is always received correctly and that the receiver knows which channel coding 200 was used in each case.

[0055] By receiving an ACK signal 211A, time-critical data 212 and non-time-critical data 213 are sent in the reverse direction as usual and protected with the weaker channel coding. However, the NACK signal 211B now signals to control unit 10 that a packet error 26 has occurred in the forward direction. Therefore, the data packet now sent from component 31 to control unit 10 is a data packet 21B.

[0056] If control unit 10 receives data packet 21B with NACK signaling 211B, it briefly interrupts the transmission of non-time-critical data 213 in the following cycle K+1 and instead uses the stronger channel coding for the next transmission in the forward direction. For the resulting data packet 210A with ACK signaling 211A, repeated packet errors 26 of the time-critical data 210A are then very unlikely.

[0057] Fig. 6, however, shows an example in which a packet error 26 occurs in the reverse direction during transmission cycle K. The procedure here is equivalent to packet errors 26 in the forward direction. The packet error 26 is detected in the control unit 10. Therefore, the data error 26 is communicated to the component 31 with the next data packet 21B in the forward direction in cycle K+1 via the NACK signaling 211B. The component 31 then interrupts the transmission of non-time-critical data 213 in cycle K+1 and instead increases the redundancy, so that a data packet 210A is sent.

[0058] If a transmission error occurs in direct succession, resulting in packet errors 26 in both the forward and reverse directions, the method works as long as the NACK signals 211B can be received without error. As a result, the stronger channel decoding would be used in the next cycle in both transmission directions, preventing further packet errors 26.

[0059] For the present method, different channel codings 200 are required, each with the same predetermined data packet length N but with different correction properties. For such an adaptive channel coding, one of the following options is conceivable, depending on the requirements: a) Convolutional codes with different puncturing, such as in WLAN (WLAN = Wireless Local Area Network). b) Use of block codes of the same block length with different higher error correction performance. It should be noted that both channel coding and channel decoding can vary significantly. This variant also requires the implementation of two different decoder algorithms. In block coding, the set of information or signaling bits is divided into blocks. Redundant check bits, called parity, are added to each block, allowing a certain number of bit errors to be corrected. The redundancy of the channel coding reduces the payload data rate while maintaining the same overall data rate. c) Iterative coding methods in which multiple codes are interleaved. One example is product codes. Each additional interleaved code increases the error correction properties but simultaneously reduces the payload rate. For the weaker channel coding, the innermost coding and the corresponding decoding could be omitted. This approach partially uses the same decoder structure for both coding variants.

[0060] Thus, in the present method, the adaptive channel coding 200 does not refer to a faulty data packet itself, but to the data packet following the faulty data packet. The individual packet error 26 is initially accepted. However, the probability of a repeat packet error 26 is reduced by the stronger channel coding using packets 210. The signaling 211B of the first packet error 26 can still occur in the first cycle. The additional delay until the following data packet is received error-free is only extended by the transmission of the additional redundancy and the somewhat more complex error correction. However, this delay is still tolerable by most real-time systems and is significantly shorter than in a method in which an attempt is made to subsequently correct the packet error 26 that has already occurred without having to retransmit the entire data packet 21, 210.Such a packet is only available error-free at the receiver after a delay resulting from error signaling, transmitting the additional redundancy, and performing error correction. Such a long delay is often incompatible with time-critical real-time systems.

[0061] A modification of the present procedure ensures that the two pieces of information explained below are always transmitted without errors during each transmission.

[0062] Firstly, it ensures that the receiver clearly and accurately knows which channel coding 200 was currently used. This eliminates the risk that either part of the strong redundancy is mistaken for non-time-critical data 213, or that non-time-critical data 213 is misinterpreted as part of the stronger redundancy and, as a result, time-critical data is incorrectly corrected.

[0063] Secondly, it ensures that the receiver can receive the correct signaling 211A, 211B without errors. If an ACK signaling 211A is misinterpreted as a NACK signaling 211B, the transmission of non-time-critical data 213 is unnecessarily interrupted in order to apply stronger channel decoding. This reduces the effective payload rate for the non-time-critical data 213 more than necessary. If, on the other hand, a NACK signaling 211A is misinterpreted as an ACK signaling 211A, the channel coding is not amplified despite an initial transmission error. This greatly increases the probability of critical multiple errors.

[0064] To avoid this, the two pieces of information can optionally be secured by a separate, strong channel coding, which is independent of the channel coding 200 of the payload data 212 and / or 213 and has an extremely low error probability. Since only two bits need to be protected here, such a permanent, stronger channel coding would be acceptable. Both pieces of information could be contained in the signaling data 211.

[0065] According to a second embodiment, the control unit 10 has absolute control over the transmission via the transmission medium 20. In this case, deterministic behavior prevails. Such behavior is typical for a network with a central controller as the control unit 10 and is common in many industrial networks. Here, the central controller as the control unit 10 has sole control over the time-critical resource allocation and the data formats used. Although for non-time-critical data 213, the slaves or the other components 31, 32, 33 can also partially access the channel of the transmission medium 20 independently, a slave cannot assume that it will receive sole access.

[0066] The sole control over the data format of the time-critical data 212 is also present in the present method, even if the component 31 in the example of Fig.5, the control unit 10, as the central controller, instructs the central controller to change the packet structure in the following cycle with the NACK signaling 211B. However, it is conceivable that the control unit 10, as the central controller, retains final control over the packet types used if this appears appropriate for certain reasons. Thus, the control unit 10, as the central controller, could, on the one hand, ignore NACK signaling 211B in the forward direction and continue to use the weaker channel coding. This is appropriate, for example, if no further time-critical data 213 is available for transmission. On the other hand, the control unit 10, as the central controller, could also use a stronger channel coding 200 in the forward direction despite the ACK signaling 211A in the reverse direction.

[0067] The channel coding 200 in the reverse direction could be controlled by the control unit 10, as the central controller, by signaling ACK and NACK 211A, 211B, regardless of whether the previous decoding was successful or not. However, components 31, 32, and 33 should not deviate from the standard procedure. They should always apply the channel coding 200 according to the signaling and report detected errors unaltered to the control unit 10, as the central controller.

[0068] A third embodiment is based on the basic assumption that only more than two packet errors 26 are critical for the correct functioning of industrial plant 1. In other words, in the case where only M≥3 consecutive errors are system-critical, the following procedure is followed. The stronger channel coding is only used when M-1 consecutive packet errors 26 have already occurred. This also results in more lead time for the network devices 10, 31, 32, 33 to change the channel coding. However, even the first packet error when using the stronger channel coding leads to a system failure.

[0069] According to a fourth embodiment, in the case where only M≥3 consecutive packet errors 26 are system-critical, the stronger channel coding is already used when fewer than M-1 consecutive packet errors 26 have occurred. Thus, a system failure only occurs when more than one packet error 26 occurs with stronger channel coding. This probability is significantly lower.

[0070] According to a fifth embodiment, in the case where only M≥3 consecutive packet errors 26 are system-critical, up to M-1 different channel coding strengths are provided and the channel coding 200 is increased for each further consecutive packet error 26.

[0071] Depending on the choice of the methods according to the first to fifth embodiments, a good compromise can always be found between usable data rate and avoidance of critical multiple errors.

[0072] All previously described embodiments of the system and the method implemented by it can be used individually or in any possible combination. In particular, all features and / or functions of the previously described embodiments can be combined as desired. In addition, the following modifications are conceivable.

[0073] The parts shown in the figures are schematic and may differ in their exact design from the forms shown in the figures, as long as their previously described functions are guaranteed.

[0074] The industrial system 1 can be a programmable logic controller (PLC). The industrial system 1 can be a CNC controller (computerized numerical control). The industrial system 1 can be or have a motion logic controller for, for example, transport systems or for guiding tools, etc.

[0075] As an alternative to the previously described method, in which the gross packet length is constant and the net data volume is temporarily reduced, e.g. by suspending non-time-critical data, it is possible that the payload data volume remains constant and the gross packet length is temporarily increased.

[0076] The method described above can be used in any configuration or system where data must be transmitted in a time-critical manner, but isolated packet errors are not critical. The approach described is particularly interesting for real-time communication in industrial automation. However, any other real-time data stream is conceivable where individual packet errors can still be corrected by interpolation, but multiple errors should be prevented. Further examples could be control tasks in vehicles (steer-by-wire, brake-by-wire), where, in the event of an error, it is more important that the current commands are transmitted successfully than that previous commands are repeated.

Claims

[1] Component (10; 31; 32; 33) for a machine or industrial plant (1), with a network communication device (100; 310; 320; 330) for controlling a transmission of data in a data network of the machine or industrial plant (1), in which data packets (21, 21A, 21B, 210, 210A) with a predetermined packet length (N) are transmitted cyclically between the component (10; 31; 32; 33) and at least one further component (31; 32; 33; 10) of the machine or the industrial plant (1), without a retransmission of time-critical data of the data packet (21, 21A, 21B, 210, 210A) being possible in a transmission cycle (K, K+1), wherein the network communication device (100; 310; 320; 330) is designed to tolerate any packet error (26) in a received data packet (21, 21A, 21B, 210, 210A) that was sent in the previous transmission cycle (K), and the at least one further component (31; 32; 33;10) in the following transmission cycle (K+1) to signal the presence or absence of a packet error (26), wherein the network communication device (100; 310; 320; 330) is designed to determine whether the at least one further component (31; 32; 33; 10) has signaled a packet error (26) for a data packet (21, 21A, 21B, 210, 210A) sent by the component (10) in the previous transmission cycle (K) or not, and wherein the network communication device (100; 310; 320; 330) is designed to adapt the number of redundancy data (215) in the data packet (21, 21A, 21B) to be sent in the following transmission cycle (K+1) depending on the determination result, which redundancy data has a correction probability of increase bit errors in the time-critical data (212) to be sent.; [2] Component (10; 31; 32; 33) according to claim 1, wherein the network communication device (100; 310; 320; 330) is designed, if no packet error (26) is detected in the preceding transmission cycle (K), to increase the number of time-critical data (212) in the data packet (21, 21A, 21B) to be sent in the following transmission cycle (K+1) or to include non-time-critical data (213) which are to be transmitted but for which a time delay is not critical, and wherein the network communication device (100; 310; 320; 330) is designed, if a packet error (26) is detected in the preceding transmission cycle (K), to increase the number of time-critical data (212) in the data packet (21, 21A, 21B) to be sent in the following transmission cycle (K+1) instead of the non-time-critical data (213) or part of the time-critical data (212) to include the redundancy data (215). [3] Component (10; 31; 32; 33) according to claim 1 or 2, wherein the network communication device (100; 310; 320; 330) is further designed to include in the data packet (21A, 21B, 210A) a first piece of information (211A, 211B) about whether a data packet (21A, 21B, 210A) that was sent in the previous cycle from the at least one further component (31; 32; 33; 10) to the component (10; 31; 32; 33) had a packet error (26) or not, and a second piece of information (200) about the current configuration of the data packet (21A, 21B, 210A) to be sent in the current transmission cycle, wherein the second piece of information is provided by the at least one further component (31; 32; 33; 10) can be used for error correction of the time-critical data (212). [4] Component (10; 31; 32; 33) according to claim 3, wherein the first information (211A, 211B) is contained in one bit, and wherein the second information (200) indicates the number of check bits in the data packet, and / or wherein the redundancy data (214; 215) are check bits. [5] Component (10; 31; 32; 33) according to one of claims 3 or 4, wherein the network communication device (100; 310; 320; 330) is further configured to examine the time-critical data (212) for bit errors and, if at least one bit error is present, to carry out an error correction of the time-critical data (212) on the basis of the second information (200) and the redundancy data (214; 215). [6] Component (10; 31; 32; 33) according to one of the preceding claims, further comprising an application (101; 311 to 31N; 321; 331 to 33N) which is designed to use the time-critical data (212), wherein the network communication device (100; 310; 320; 330) is further designed to transfer the time-critical data (212) to the application (101; 311 to 31N; 321; 331 to 33N) if the network communication device (100; 310; 320; 330) has not detected a packet error (26), and wherein the network communication device (100; 310; 320; 330) is further designed to provide the application (101; 311 to 31N; 321; 331 to 33N) that transmission errors have occurred and no valid data is available in this transmission cycle (K; K+1) when the network communication device (100; 310; 320; 330) has detected a packet error (26). [7] Component (10) according to one of the preceding claims, wherein the component is a control unit (10) which is designed to control at least one tool of the industrial plant (1) as the at least one further component (31; 32; 33), or wherein the component is a control unit (10) of a machine, wherein the control unit (10) is designed to control at least one sensor and / or actuator as the at least one further component (31; 32; 33), or wherein the component is a control unit (10) of a vehicle, wherein the control unit (10) is designed to control a steering power assistance or a brake power boost of the vehicle. [8] Data network, with a wired or wireless transmission medium (20), and at least two components (10; 31; 32; 33) according to one of the preceding claims, wherein the at least two components (10; 31; 32; 33) are connected to one another via the transmission medium (20) in order to be able to carry out a transmission of data in the data network, in which data packets (21, 21A, 21B, 210, 210A) with a predetermined packet length (N) are transmitted cyclically between one of the components (10) and at least one further component (31; 32; 33) of the machine or industrial plant (1), without a retransmission of time-critical data of the data packet (21, 21A, 21B, 210, 210A) being possible in a transmission cycle (K, K+1). [9] Data network according to claim 8, wherein a control device (10) is designed to control at least one tool of the industrial plant (1) as the further component (31; 32; 33) and / or to control at least one sensor (331 to 33N) and / or actuator (311 to 311N) as the further component (31; 32; 33), wherein the control device (10) is designed such that the control device (10) as the central controller can exchange data packets (21, 21A, 21B, 210, 210A) bidirectionally with each further component, and wherein each further component (31; 32; 33) is designed such that each of the further components (31; 32; 33) can exchange data packets (21, 21A, 21B, 210, 210A). [10] Method for controlling at least one component (10; 31; 32; 33) in a machine or industrial plant (1), wherein a network communication device (100; 310; 320; 330) for controlling a transmission of data in a data network of the machine or industrial plant (1) is provided for the at least one component (10; 31; 32; 33), in which network communication device data packets (21, 21A, 21B, 210, 210A) with a predetermined packet length (N) are transmitted cyclically between the component (10; 31; 32; 33) and at least one further component (31; 32; 33; 10) of the machine or industrial plant (1), without a transmission repetition of time-critical data of the data packet (21, 21A, 21B, 210, 210A) is possible, the method comprising the steps of tolerating, with the network communication device (100; 310; 320;330), of a possible packet error (26) in a received data packet (21, 21A, 21B, 210, 210A) that was sent in the previous transmission cycle (K), determining, with the network communication device (100; 310; 320; 330), whether the at least one further component (31; 32; 33; 10) has signaled a packet error (26) for a data packet (21, 21A, 21B, 210, 210A) sent by the component (10) in the previous transmission cycle (K) or not, adapting, with the network communication device (100; 310; 320; 330), depending on the determination result, in the data packet (21, 21A, 21B) to be sent in the following transmission cycle (K+1) the number of redundancy data (215) which increase a probability of correcting bit errors in the time-critical data (212) to be transmitted, and including signaling data (211A, 211B) in the data packet (21A, 21B, 210A) to be transmitted which is assigned to the at least one further component (31; 32; 33;10) signal the presence or absence of a packet error (26) in the following transmission cycle (K+1);

Citation Information

Patent Citations

  • Techniques for a hybrid automatic repeat request (HARQ) mechanism with polar codes

    EP3455975B1

  • Error management

    US20110185262A1

  • Fast and reliable wireless communication

    US20110225479A1

  • Redundant encoding

    US20150169396A1