JP5445233B2 - Robot controller - Google Patents

Description

  The present invention relates to a robot controller that controls the operation of a robot in accordance with a control command given from the outside.

  A controller for controlling the robot performs drive control of each axis of the robot, I / O switching control, and the like according to a control command (command) given from the outside. This command is transmitted from, for example, a personal computer based on a program arbitrarily created by the user. The command is transmitted in units of packets to which control information (header, footer, serial number, etc.) is added. The data structure of this packet is predetermined. The user himself debugs the program. Therefore, when the user makes a mistake in creating the program, the transmitted packet may not have a regular data structure. On the other hand, if the robot performs an operation unintended by the user, it may cause a dangerous situation. For this reason, it is necessary to take countermeasures on the controller side so that the above situation is not caused by the user's program.

  For this reason, if the controller cannot receive the command properly, such as when the received packet does not have a regular data structure, the command may have an incorrect data length. Since there is, the packet and all subsequent packets are discarded (reset). Thereafter, the same packet is transmitted again from the personal computer side, and the receiving operation is performed again on the controller side. As a result, even if a packet with an incorrect structure is transmitted, the controller does not respond to this, so that it is possible to prevent a situation where the robot performs an operation not intended by the user.

  As described above, for example, Patent Document 1 discloses a technique for performing a reset when a problem relating to data reception occurs. That is, Patent Document 1 discloses a technique for resetting and transmitting / receiving data again when an abnormality occurs in non-contact input means for inputting / outputting data without contact with the outside. Yes.

JP-A-5-143791

  The above-described command receiving method may cause the following problems. In other words, if packets are continuously transmitted from the personal computer and one of them has a problem (such as an irregular data structure), all of the continuously transmitted packets are reset. For this reason, the user cannot recognize which packet has a problem and has been reset, and may repeat the same mistake.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a robot controller that can identify where the problem is when a control command cannot be properly received. .

  According to the first aspect of the present invention, the control unit receives a control command given from the outside in units of packets. The control unit sequentially stores the received packets from the start address to the end address of the packet storage memory in the order received. This packet has a predetermined data structure, and is composed of a fixed length header, a message length, a serial number and a footer, and a variable length data portion. The header is fixed value data provided at the beginning of the packet. The message length is data that is provided immediately after the header and indicates the total data length of the packet. The serial number is data that is a unique value for each packet. The data part is data indicating control contents. The footer is fixed value data provided at the end of the packet. The control unit determines whether or not the packet stored in the packet storage memory is normal, that is, whether or not it has a normal data structure as described above. And a control part performs operation control of a robot based on the control content which the data part of the packet judged to be normal shows.

  The control unit performs the following process when determining whether or not the packet is normal. That is, when the packet is stored in the packet storage memory, the control unit executes the first confirmation process with the last address among the addresses where the packet is stored as the reference address. In the first confirmation process, the control unit searches for a header from the reference address toward the start address (header search process). Then, when the header is found, the control unit acquires the total data length indicated by the message length provided immediately after the header (total data length acquisition process). In addition, the control unit detects the footer address where the footer is stored using the header address where the found header is stored and the acquired total data length (footer address detection process). Specifically, the footer address is obtained by adding the total data length to the header address. The header address mentioned here may be the first address (start address) occupied by the header or the last address (end address). Alternatively, each address from the start to the end occupied by the header (that is, each address of the memory area occupied by the header itself = the amount of addresses) may be used. Also, the footer address represents the same as the header address. Then, the control unit determines whether or not a footer exists at the detected footer address (footer confirmation processing).

  If it is determined in the footer confirmation process that there is no footer, the packet including the footer is abnormal. In this case, the control unit performs the header search process again from the header address of the found header toward the start address. That is, the control unit re-executes the header search process to search for the header of the next packet. On the other hand, when it is determined in the footer confirmation process that a footer exists, the packet including the footer is normal. In this case, the control unit executes a second confirmation process.

  In the second confirmation process, the control unit confirms whether or not the data length from the reference address to the head address of the header address matches the acquired total data length. The reason for performing such second confirmation processing is as follows. That is, if the packets having normal data structures (normal packets) are continuous, the data lengths match. On the other hand, when a normal packet (for example, no header, no footer, wrong data length, etc.) is received after an abnormal packet, the data lengths do not match.

  For this reason, when it is determined that the data lengths match in the second confirmation process, the control unit determines that the packet including the footer confirmed by the footer confirmation process is normal, and stores the serial number. (Normal data storage processing). In addition, when it is determined in the second confirmation process that the data does not match, the control unit stores a packet stored after the packet including the footer confirmed by the footer confirmation process (confirmed footer). The packet stored next to the end address side of the packet including) is abnormal, and stores abnormal data indicating that (abnormal data storage processing). At this time, the control unit stores the abnormal data for identifying the packet determined to be abnormal using the serial number of the packet including the confirmed footer. For example, the abnormal data has contents such as “the packet (serial number: unknown) stored immediately after the packet (serial number: XXXX) has an abnormality”.

  When the normal data storage process is executed, it is determined that the packet has a normal data structure. In this case, it is confirmed whether or not the next packet (packet stored next to the start address side of the packet determined to be normal) is normal. That is, when the normal data storage process is executed, the control unit uses the address advanced by one address (for example, 1 byte) from the head address of the header of the packet including the stored serial number toward the start address side as the reference address. The first confirmation process is executed above. On the other hand, when the abnormal data storage process is executed, since it is confirmed that an abnormal packet exists, it is not possible to check whether or not the subsequent packets are normal. Therefore, when the abnormal data storage process is executed, the control unit outputs the serial number and the abnormal data stored at that time to the outside (notification process).

  If no abnormality is found in all the packets as a result of repeating the above processes, the start address is exceeded in the header search process of the first confirmation process. In this case, the control unit performs notification processing for outputting the stored serial number to the outside. In this case, only an arbitrary serial number (for example, the last stored serial number) may be output, or all serial numbers may be output. Thus, the header search process is terminated with the start address as the end point. If the start address (for example, address 0000) is set as the end point, there is no address that is earlier (the younger number) than the start address, so that the header search process is surely terminated in any situation. Become. Therefore, the possibility of falling into an infinite loop is extremely low.

  In this way, the control unit confirms whether or not the received packet is normal, and as a result, outputs the serial number of the packet determined to be normal to the outside and is abnormal. Abnormal data for specifying the determined packet is output to the outside. According to such a configuration, when a control command reception error occurs, a user who creates a control command (program) has an error in which packet based on the serial number and error data output from the controller side. You can easily check if The user can accurately and efficiently debug the program based on this information, and a situation in which a similar reception error occurs thereafter is prevented.

  As described above, the control unit is normal in order from the last packet stored closest to the final address of the packet storage memory to the first packet stored closest to the start address. Confirm whether or not. When a new packet is received during such confirmation, the newly received packet is stored after the last packet at that time. For this reason, the packet is confirmed only for the packet stored at the start of the confirmation. Therefore, the confirmation as to whether or not the packet is normal ends as soon as the confirmation of the packet that has been initially confirmed is completed, regardless of whether or not a new packet is received during the confirmation. Then, at the time of completion, it is possible to execute control based on the control content indicated by the data portion of the packet that has been confirmed. According to such a configuration, the delay time from when a packet is received until the robot operation control based on the packet is started can be shortened as much as possible.

  According to the means described in claim 2, when it is determined in the second confirmation process that the data length does not match the total data length, the control unit determines that the data length is larger than the total data length. Performs the abnormal data storage processing in the same manner as the means described in claim 1. If the same data is accidentally present in portions other than the footer and header of the packet, the data length in the second confirmation process becomes smaller than the total data length. In this means, assuming such a rare case, in the second confirmation process, if the data length is smaller than the total data length, the header search from the header address of the found header toward the start address is performed. Execute the process. That is, since the header found at this time is not correct, the header search process is continued until a correct header is found. According to such a configuration, it is possible to accurately determine whether or not a packet is normal even in the rare case described above.

The figure which shows schematically the structure of the robot system which shows one Embodiment of this invention. Diagram showing data structure of packet Diagram for explaining the flow of reception processing Flow chart showing the flow of packet confirmation processing Diagram showing a case where a packet with no header is received The figure which shows the case where the packet which does not have the footer is received Figure showing a case where a packet with incorrect message length is received The figure which shows the case where packet confirmation processing is done in order from the first packet The figure which shows the case where packet confirmation processing is done in order from the last packet

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 shows a system configuration of a general industrial robot. A robot system 1 shown in FIG. 1 includes a robot 2 and a controller 3 that controls the robot 2. The robot 2 is a robot having an arbitrary configuration, for example, for assembling parts or for inspecting parts. The controller 3 is connected to a teaching pendant 4 and a personal computer 5 (hereinafter referred to as a personal computer 5) as peripheral devices. Note that the personal computer 5 is connected as necessary, and is not always connected.

  The robot 2 is configured as, for example, a 6-axis vertical articulated robot. The robot 2 includes a base 6, a shoulder portion 7 that is rotatably supported by the base 6 in the horizontal direction, a lower arm 8 that is rotatably supported by the shoulder portion 7 in the vertical direction, and the lower arm 8. A first upper arm 9 rotatably supported in the vertical direction, a second upper arm 10 rotatably supported by the first upper arm 9, and a vertical direction supported by the second upper arm 10 The wrist 11 is rotatably supported on the wrist 11, and the flange 12 is rotatably supported on the wrist 11 by twisting.

  The base 6, the shoulder portion 7, the lower arm 8, the first upper arm 9, the second upper arm 10, the wrist 11 and the flange 12 function as the arm of the robot 2. Although not, an end effector (hand) is attached. A plurality of axes provided in the robot 2 are driven by servo motors (not shown) provided corresponding to the respective axes. The robot 2 and the controller 3 are connected by a connection cable 13. Thus, the controller 3 controls the drive of the servo motor that drives each axis of the robot 2.

  The teaching pendant 4 is, for example, a size that can be operated by being carried by a user or carried by a hand, and is formed in, for example, a thin, substantially rectangular box shape. The teaching pendant 4 is provided with various key switches 14, and the user performs various input operations using the key switches 14. The teaching pendant 4 is connected to the controller 3 via a cable 15 and executes high-speed data transfer with the controller 3 via a communication interface. Information such as an operation signal input by operating the key switch 14 is transmitted from the teaching pendant 4 to the controller 3. The controller 3 supplies driving power to the teaching pendant 4 together with a control signal and a display signal.

  The user can execute various functions such as operation and setting of the robot 2 using the teaching pendant 4 described above. For example, by operating the key switch 14, a user can call a control program stored in advance, The robot 2 can be activated and various parameters can be set. Also, various teaching operations can be executed by operating the robot 2 by manual operation.

  The personal computer 5 is, for example, a general-purpose notebook personal computer. The user can use the personal computer 5 to create a program in which control commands (commands) for performing drive control of each axis of the robot 2 and I / O switching control are described. The personal computer 5 is connected to the controller 3 via the cable 16. Thereby, the personal computer 5 can transmit various commands based on the created program to the controller 3. This command is transmitted in packet units having a predetermined data structure.

  When the controller 3 (corresponding to the controller of the robot) receives a packet transmitted from the personal computer 5, it performs packet confirmation processing and reception processing (details will be described later). After performing these processes, the controller 3 interprets the content (command) of data included in a packet having a normal data structure among the received packets, and controls the operation of the robot 2 based on the content.

  FIG. 2 shows a regular data structure of a packet in this embodiment. As shown in FIG. 2, the packet P is composed of a header H, a message length M, a serial number S, a data part D, and a footer F in this order. The header H is data of a fixed length (for example, 1 byte), and has a value specific to the data structure of the present embodiment. The message length M is fixed-length data (for example, 4 bytes). The message length M is data indicating the length (number of bytes) of the entire packet P (from the first bit of the header H to the last bit of the footer F).

  The serial number S is data having a fixed length (for example, 4 bytes). The serial number S is data indicating a serial number and has a unique value for each packet P. The data part D is variable length data (for example, 0 to 12 bytes). The data part D is data indicating a command (control instruction) for instructing the operation of the robot 2 or the like. The footer F is data of a fixed length (1 byte in the present embodiment), and has a value specific to the data structure of the present embodiment. The packet P having the structure shown in FIG. 2 has a normal data structure, and a packet having a data structure other than this has a non-normal data structure.

  FIG. 3 shows a configuration for receiving a packet transmitted from a personal computer. In FIG. 3, (a) shows a state immediately after receiving a packet, and (b) shows a state where reception processing for one packet is completed. As shown in FIG. 3A, the packets P <b> 1 to P <b> 4 transmitted from the personal computer 5 are temporarily stored in the packet storage memory 21 of the controller 3. At this time, the packets P1 to P4 are stored in the packet storage memory 21 in the order received. In other words, the received packets P1 to P4 are sequentially stored from the start address side to the end address side of the packet storage memory 21 in that order.

  The control unit 22 is configured mainly with a microcomputer, for example, and controls the operation of the robot 2 based on a command transmitted from the personal computer 5. When a packet is stored in the packet storage memory 21, the control unit 22 performs a packet confirmation process on all the stored packets (details will be described later). When determining that there is no abnormality in the packet confirmation process, the control unit 22 performs a reception process on the packet stored closest to the start address, that is, the first packet P1. On the other hand, when it is determined that “abnormality exists” in the packet confirmation process, the control unit 22 transmits to the personal computer 5 data indicating which packet is abnormal, as will be described in detail later.

  Now, in the reception process, the control unit 22 analyzes the content of the data portion D1 of the packet P1, and executes a predetermined operation according to the control content (command) obtained by the analysis. The packet P1 for which the reception process has been completed is erased from the packet storage memory 21, and the next packet P2 is moved to the start address side. That is, as shown in FIG. 3B, after the reception process of the packet P1 is completed, the next packet P2 is stored in a place closest to the start address. Thereafter, reception processing for the packet P2 is performed, and thereafter, similar processing is repeatedly performed until reception processing for all stored packets is completed.

Next, the contents of the packet confirmation process will be described with reference to FIGS.
FIG. 4 is a flowchart showing the contents of the packet confirmation processing by the control unit 22. First, the flow of packet confirmation processing when the received packet has a regular data structure as shown in FIG. When the packet confirmation process is started, the address at which the last bit of the footer F4 of the packet P4, which is the address closest to the end address among the addresses where the packets P1 to P4 are stored, is stored. (Step A1). Here, “move” means that the control unit 22 changes the address that is read from the packet storage memory 21 and checked. In the subsequent step A2, the destination address at this point is stored as a reference address. In this case, the address where the last bit of the footer F4 is stored is stored as the reference address.

  In the subsequent step A3, the packet is traced back from the last address by the minimum data size of the packet. That is, it moves to the start address side by the minimum data size of the packet P in this embodiment from the last address. The minimum data size of the packet P includes all the data length of the header H (fixed), the data length of the message length M (fixed), the data length of the data part D (variable), and the data length of the footer F (fixed). It is an addition. However, the data length of the data part D is assumed to be the smallest possible value (for example, 0 bytes). In the subsequent step A4, it is determined whether or not the moved destination address (the retroactive address) has exceeded the start address of the packet storage memory 21.

  In this case, since all received packets have a regular data structure, the start address is not exceeded, and “NO” is determined in the step A4, and the process proceeds to the step A5. In step A5, it is determined whether or not the data starting from the destination address is a header. If the data part D does not particularly represent a command and the data length is 0 byte, the data starting from the destination address is a header (YES), so the process proceeds to step A7. On the other hand, if the data part D represents some command and the data length exceeds 0 bytes, the data starting from the destination address is not a header (NO), so the process proceeds to step A6. In step A6, the program further moves to the start address side by 1 byte and returns to step A4. Thereafter, steps A4 to A6 are repeatedly executed until the data starting from the destination address becomes the header (until “YES” in step A5), that is, until the header is found.

  As a result of the repeated execution of steps A4 to A6, if it is confirmed that the data starting from the destination address is a header ("YES" in step A5), the process proceeds to step A7. In step A7, data of the message length M4 located after the header H4 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F4 (corresponding to the footer address) is obtained from the destination address (the address where the header H4 is stored and corresponding to the header address) and the data length of the packet P4 represented by the message length M4. The process proceeds to step A9. In step A9, it is determined whether or not the data stored at the address obtained in step A8 is actually a footer. In this case, since all packets have a regular data structure, the obtained address matches the actual footer F4 address. Accordingly, since the data stored at the obtained address is the footer F4 (YES), the process proceeds to step A10.

  In step A10, it is determined whether or not the movement amount (data length) from the reference address obtained in step A2 to the start address of the header H4 obtained in step A5 matches the data length represented by the message length M4. . In this case, since all packets have a regular data structure, the amount of movement matches the data length (YES), the process proceeds to step A11. In step A11, it is determined that the packet P4 is normal (has a regular data structure), the serial number S4 of the packet P4 is stored, and the process proceeds to step A12.

  In step A12, the program further moves to the start address side by 1 byte and returns to step A2. In this case, the process moves to the address where the last bit of the footer F3 of the packet P3, which is the packet immediately before the packet P4, is stored, and returns to step A2. Thereafter, it is confirmed whether or not there is an abnormality in the packets P3, P2, and P1 by the same procedure as that of the packet P1 described above (steps A2 to A12). Then, after it is confirmed that the packet P1 is normal, the destination address exceeds the start address of the packet storage memory 21 when going back one more byte in step A12. Accordingly, “YES” is determined in the step A4, and the process proceeds to the step A13. In step A13, a notification process is performed, and then the process ends. In this notification process, the serial numbers S4 to S1 of the packets P4 to P1 determined to be normal are transmitted to the personal computer 5 together with data indicating that the packets are normally received. In this notification process, only an arbitrary serial number among the serial numbers S4 to S1 may be transmitted. That is, for example, only the serial number S1 of the first packet P1 or the serial number S4 of the last packet P4 may be transmitted.

Next, the flow of packet confirmation processing when the received packet does not have a regular data structure will be described. Here, the following three patterns will be described.
(1) When there is no header FIG. 5 shows a pattern in which the header (H2) does not exist in the second received packet P2. The flow of packet confirmation processing when receiving three packets P1 to P3 including a packet P2 having no header (H2) as shown in FIG. 5 is as follows. That is, the packet confirmation process of FIG. 4 is started, and as a result of first executing steps A1 to A10, it is confirmed that the packet P3 stored at the location closest to the final address of the packet storage memory 21 is normal. (Both “YES” in steps A9 and A10).

  In the subsequent step A11, the serial number S3 of the packet P3 is stored, and the process proceeds to step A12. In step A12, the process moves to the address where the last bit of the footer F2 of the packet P2, which is the packet immediately before the packet P3, is stored, and the process returns to step A2. In step A2, the destination address at this point is stored as a reference address. In this case, the address where the last bit of the footer F2 is stored is stored as the reference address, and the process proceeds to Step A3.

  In step A3, the process goes back by the minimum data size of the packet and proceeds to step A4. Thereafter, steps A4 to A6 are repeated until it is confirmed that the data starting from the destination address is a header. In this case, however, the packet P2 has no header. For this reason, it is confirmed that the data starting from the destination address is the header when the data has moved to the address (= start address) where the first bit of the header H1 of the packet P1 is stored. That is, although it should have been searching for the header of the packet P2, it has found the header of the packet P1.

  Thus, when it is confirmed that the data starting from the destination address is the header H1 (“YES” in step A5), the process proceeds to step A7. In step A7, data of the message length M1 located after the header H1 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F1 is obtained from the destination address (address where the header H1 is stored) and the data length of the packet P1 represented by the message length M1, and the process proceeds to step A9.

  In step A9, it is determined whether or not the data stored at the address obtained in step A8 is actually a footer. In this case, the footer F1 of the packet P1 is stored at the address obtained in step A8. Accordingly, it is determined that the footer is stored at the address obtained in step A8 (YES), and the process proceeds to step A10.

  In step A10, it is determined whether or not the movement amount (data length) from the reference address stored in step A2 to the start address of the header H1 obtained in step A5 matches the data length represented by the message length M1. The In this case, the movement amount is the data length from the address where the last bit of the footer F2 is stored to the address where the first bit of the header H1 is stored. That is, the movement amount is a length obtained by adding the data length of the entire packet P2 and the data length of the entire packet P1. On the other hand, the data length represented by the message length M1 is the data length of the entire packet P1. Therefore, it is determined that the amount of movement does not match the data length (“NO” in step A10), and the process proceeds to step A14.

  In step A14, it is determined whether the amount of movement is longer than the data length represented by the message length M1. In this case, it is determined that the moving amount is longer than the data length represented by the message length M1 (YES), and the process proceeds to step A15. In step A15, it is determined that the packet P1 including the message length M1 is normal and the packet (P2) immediately before the packet P1 is abnormal. At this time, the serial number S1 of the packet P1 is stored. Then, after the notification process in step A13 is performed, the process ends. In this notification process, the serial numbers S3 and S1 of the packets P3 and P1 are transmitted to the personal computer 5 together with data indicating that the packets are normally received. At this time, abnormal data indicating that the packet (P2) received immediately after the packet P1 is abnormal is also transmitted to the personal computer 5.

(2) When no footer exists FIG. 6 shows a pattern in which the footer F2 does not exist in the second received packet P2. The flow of the packet confirmation process when receiving three packets P1 to P3 including the packet P2 without the footer F2 as shown in FIG. 6 is as follows. That is, the packet confirmation process of FIG. 4 is started, and as a result of first executing steps A1 to A10, it is confirmed that the packet P3 stored at the location closest to the final address of the packet storage memory 21 is normal. (Both “YES” in steps A9 and A10).

  In the subsequent step A11, the serial number S3 of the packet P3 is stored, and the process proceeds to step A12. In step A12, the process moves to the address where the last bit of the data part D2 of the packet P2, which is the packet immediately before the packet P3, is stored, and the process returns to step A2. In step A2, the address where the last bit of the data part D2 is stored is stored as a reference address, and the process proceeds to step A3.

  In step A3, the process goes back by the minimum data size of the packet and proceeds to step A4. Thereafter, steps A4 to A6 are repeated until it is confirmed that the data starting from the destination address is a header. As a result, if it is confirmed that the data starting from the destination address is the header H2 (“YES” in step A5), the process proceeds to step A7. In step A7, data of the message length M2 located after the header H2 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F2 is obtained from the address where the header H2 is stored and the data length of the packet P2 represented by the message length M2, and the process proceeds to step A9.

  In step A9, it is determined whether or not the data stored at the address obtained in step A8 is actually a footer. In this case, there is no footer in the packet P2, and a part of the data part D2 is stored at the address obtained in step A8. Therefore, it is determined that the footer is not stored at the address obtained in step A8 (NO), and the process proceeds to step A6. In step A6, the process moves to the address where the last bit of the footer F1 of the packet P1, which is the packet immediately before the packet P2, is stored, and returns to step A4. Thereafter, steps A4 to A6 are repeatedly executed until it is confirmed that the data starting from the destination address is a header, that is, until the next header is found.

  As a result of the repeated execution of steps A4 to A6, if it is confirmed that the data starting from the destination address is the header H1 (“YES” in step A5), the process proceeds to step A7. In step A7, data of the message length M1 located after the header H1 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F1 is obtained from the address where the header H1 is stored and the data length of the packet P1 represented by the message length M1, and the process proceeds to step A9. In this case, the footer F1 of the packet P1 is stored at the address obtained in step A8. Accordingly, in step A9, it is determined that the footer is stored at the address obtained in step A8 (YES), and the process proceeds to step A10.

  In Step A10, it is determined whether or not the movement amount (data length) from the reference address stored in Step A2 to the start address of the header H1 obtained in Step A5 matches the data length represented by the message length M1. . In this case, the amount of movement is the data length from the address where the last bit of the data part D2 is stored to the address where the first bit of the header H1 is stored. That is, the movement amount is a length obtained by adding the data length of the entire packet P2 and the data length of the entire packet P1. On the other hand, the data length represented by the message length M1 is the data length of the entire packet P1. Therefore, it is determined that the amount of movement does not match the data length (“NO” in step A10), and the process proceeds to step A14.

  In step A14, it is determined that the amount of movement is longer than the data length represented by the message length M1 (YES), and the process proceeds to step A15. In step A15, it is determined that the packet P1 including the message length M1 is normal and the packet (P2) immediately before the packet P1 is abnormal. At this time, the serial number S1 of the packet P1 is stored. Then, after the notification process in step A13 is performed, the process ends. In this notification process, the serial numbers S3 and S1 of the packets P3 and P1 are transmitted to the personal computer 5 together with data indicating that the packets are normally received. At this time, abnormal data indicating that the packet (P2) received immediately after the packet P1 is abnormal is also transmitted to the personal computer 5.

(3) When the message lengths do not match FIG. 7 shows a pattern in which the message length M2 of the second received packet P2 is incorrect data (data that does not correctly represent the data length of the entire packet P2). The flow of packet confirmation processing when receiving three packets P1 to P3 including a packet P2 whose message length M2 is incorrect data as shown in FIG. 7 is as follows. That is, the packet confirmation process of FIG. 4 is started, and as a result of first executing steps A1 to A10, it is confirmed that the packet P3 stored at the location closest to the final address of the packet storage memory 21 is normal. (Both “YES” in steps A9 and A10).

  In the subsequent step A11, the serial number S3 of the packet P3 is stored, and the process proceeds to step A12. In step A12, the process moves to the address where the last bit of the footer F2 of the packet P2, which is the packet immediately before the packet P3, is stored, and the process returns to step A2. In step A2, the address where the last bit of the footer F2 is stored is stored as the reference address, and the process proceeds to step A3.

  In step A3, the process goes back by the minimum data size of the packet and proceeds to step A4. Thereafter, steps A4 to A6 are repeated until it is confirmed that the data starting from the destination address is a header. As a result, if it is confirmed that the data starting from the destination address is the header H2 (“YES” in step A5), the process proceeds to step A7. In step A7, data of the message length M2 located after the header H2 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F2 is obtained from the address where the header H2 is stored and the data length of the packet P2 represented by the message length M2, and the process proceeds to step A9.

  In step A9, it is determined whether or not the data stored at the address obtained in step A8 is actually a footer. In this case, since the message length M2 is incorrect data, the footer F2 is not stored in the address obtained in step A8. Therefore, it is determined that the footer is not stored at the address obtained in step A8 (NO), and the process proceeds to step A6. In step A6, the process moves to the address where the last bit of the footer F1 of the packet P1, which is the packet immediately before the packet P2, is stored, and returns to step A4. Thereafter, steps A4 to A6 are repeatedly executed until it is confirmed that the data starting from the destination address is a header, that is, until the next header is found.

  As a result of the repeated execution of steps A4 to A6, if it is confirmed that the data starting from the destination address is the header H1 (“YES” in step A5), the process proceeds to step A7. In step A7, data of the message length M1 located after the header H1 is acquired, and the process proceeds to step A8. In step A8, the address of the footer F1 is obtained from the address where the header H1 is stored and the data length of the packet P1 represented by the message length M1, and the process proceeds to step A9. In this case, the footer F1 of the packet P1 is stored at the address obtained in step A8. Accordingly, in step A9, it is determined that the footer is stored at the address obtained in step A8 (YES), and the process proceeds to step A10.

  In step A10, it is determined whether or not the movement amount (data length) from the reference address stored in step A2 to the start address of the header H1 obtained in step A5 matches the data length represented by the message length M1. The In this case, the amount of movement is the data length from the address where the last bit of the data part D2 is stored to the address where the first bit of the header H1 is stored. That is, the movement amount is a length obtained by adding the data length of the entire packet P2 and the data length of the entire packet P1. On the other hand, the data length represented by the message length M1 is the data length of the entire packet P1. Therefore, it is determined that the amount of movement does not match the data length (“NO” in step A10), and the process proceeds to step A14.

  In step A14, it is determined that the amount of movement is longer than the data length represented by the message length M1 (YES), and the process proceeds to step A15. In step A15, it is determined that the packet P1 including the message length M1 is normal and the packet (P2) immediately before the packet P1 is abnormal. At this time, the serial number S1 of the packet P1 is stored. Then, after the notification process in step A13 is performed, the process ends. In this notification process, the serial numbers S3 and S1 of the packets P3 and P1 are transmitted to the personal computer 5 together with data indicating that the packets are normally received. At this time, abnormal data indicating that the packet (P2) received immediately after the packet P1 is abnormal is also transmitted to the personal computer 5.

  In any of the above cases (1) to (3) and the case where all packets have a regular data structure, if the same data as the header is present in the data portion, for example, step A5 There is a possibility that data that is not a header is erroneously determined to be a header in the determination. However, in such a case, since there is usually no message length behind the erroneously determined header, “NO” is determined in the determination of step A9. Thus, steps A4 to A6 are repeatedly executed until a regular header is found.

  However, if the data after the erroneously determined header is accidentally the data having the same format as the data length, and the footer is accidentally present at the address obtained from these in step A8, the erroneously determined header May determine whether the data structure of the packet is correct. However, in such a case, since the amount of movement from the reference address to the erroneously determined header start address does not match the data length represented by the message length, the determination in step A10 is always “NO”. Further, since the movement amount is shorter than the data length represented by the message length, “NO” is determined in the subsequent determination of Step A11. Thus, steps A4 to A6 are repeatedly executed until a regular header is found.

  In this embodiment, steps A2 to A6 correspond to header search processing, step A7 corresponds to total data length acquisition processing, step A8 corresponds to footer address detection processing, and step A9 corresponds to footer confirmation processing. To do. Steps A2 to A9 correspond to the first confirmation process, Steps A10 and A14 correspond to the second confirmation process, Step A11 corresponds to the normal data storage process, and Step A15 corresponds to the abnormal data storage process. Step A13 corresponds to notification processing.

As described above, according to the present embodiment, the following effects can be obtained.
Packets transmitted from the personal computer 5 are temporarily stored in the packet storage memory 21 of the controller 3. The control unit 22 of the controller 3 executes packet confirmation processing for all packets stored in the packet storage memory 21. In this packet confirmation process, each packet is individually judged as normal or abnormal. Then, the control unit 22 transmits the serial number of the packet determined to be normal to the personal computer 5 together with data indicating that the packet has been normally received. For a packet that is determined not to be normal, its serial number cannot be acquired, and even if it can be acquired, it is unknown whether it is correct. For this reason, the control unit 22 identifies the packet determined to be not normal, for example, in the form of a packet received immediately after the packet determined to be normal, and transmits abnormal data indicating that fact to the personal computer 5.

  As described above, in this embodiment, when the command transmitted from the personal computer 5 cannot be properly received, the data indicating where the problem occurred in the packet constituting the command (which packet was abnormal) is displayed. It is transmitted to the personal computer 5. As a result, when a reception error occurs in the transmitted command, the user operating the personal computer 5 can easily grasp which part of the transmitted command has a problem, and thus the program Can be easily debugged.

  When performing the packet confirmation process, the control unit 22 confirms whether or not the packet is normal in order from the last packet stored at a location closest to the final address of the packet storage memory 21. The reason for this will be described in comparison with a case where packet confirmation processing is performed in order from the first packet (packet stored closest to the start address) (hereinafter referred to as a comparative example). FIG. 8 is a diagram illustrating a state of the packet storage memory 21 when the packet confirmation process of the comparative example is performed. FIG. 9 is a diagram illustrating a state of the packet storage memory 21 when the packet confirmation process of the present embodiment is performed.

  In the comparative example, as shown in FIG. 8A, when the packets P1 and P2 are received in this order, the check is performed from the packet P1. During this check, when a new packet P3 is received as shown in FIG. 8B, the packet P3 is stored after the packet P2. For this reason, this packet confirmation processing is continued until the check of the packet P3 is completed. Further, during this check, as shown in FIG. 8C, when a new packet P4 is received, the packet P4 is stored after the packet P3, and the packet confirmation process continues until the check of the packet P4 is completed. Will do. Therefore, in the case of the comparative example, when a new packet is received while the packet confirmation process for a predetermined packet is being performed, the end of the packet confirmation process is extended each time. In such a case, there is a possibility that it takes a long time from when a packet is received until reception processing for that packet is executed. That is, there is a possibility that the delay time from the reception of the packet to the start of the operation control of the robot 2 based on the control content represented by the data portion of the packet may be increased.

  On the other hand, in the present embodiment, as shown in FIG. 9A, when packets P1 and P2 are received in this order, whether or not the packet P2 is normal is checked (hereinafter referred to as a check). Is called. During this check, when new packets P3 and P4 are received in this order as shown in FIG. 9B, the packets are stored in the order of packets P3 and P4 after the packet P2. In this case, in the packet confirmation processing of this embodiment, the packet P2 is continuously checked without being affected by the newly stored packets P3 and P4, and then the packet P1 is checked.

  In this way, after the check of the packets P1 and P2 is completed, at that time, among the packets stored in the packet storage memory 21, the check is performed on the packets P3 and P4 that have not been checked yet. It will be. Therefore, in the packet confirmation process of the present embodiment, when the packet confirmation process for a predetermined packet is executed, the packet confirmation process is temporarily ended as soon as the check for the packet is completed. Therefore, it is possible to execute reception processing for a packet for which packet confirmation processing has been completed at this point. According to such a configuration, the robot based on the control content represented by the data portion of the packet from the reception of the packet without receiving a long time from the reception of the packet until the reception processing for the packet is executed. The delay time until the second operation control is started can be shortened as much as possible.

  In addition, when executing the header search process, after the reference address is stored, it is configured to move to the start address side as much as the minimum data size of the packet assumed from the reference address. When moving from the reference address (the last address of the packet) to the start address side by the minimum data size of the packet, if the data size of the packet is the minimum data size, a header exists at the destination address. If the data size of the packet is larger than the minimum data size, the destination address has data (message length, serial number, data portion) after the header. That is, in any case, the packet header is not missed. Furthermore, with this configuration, the load on the control unit 22 required to find the header can be reduced as compared with a configuration in which the header is searched by moving one byte at a time from the reference address to the start address side. Processing time can be shortened.

  Now, let us consider a case in which two packets that do not have a regular data structure are received in succession, and when these two are combined, they are regarded as one packet having a regular data structure. . For example, when a packet Pb having only a data portion and a footer is received after a packet Pa having only a header, a message length, and a serial number, the data length indicated by the message length of the packet Pa is the data length of the entire packet Pa. If it is the same as the sum of the data length of the entire packet Pb, the result is the same as receiving a packet having one regular data structure. In this case, the configuration of the present embodiment cannot detect that there is an abnormality in each packet Pa and Pb. However, this is not a problem because the above cases are very rare and usually rarely occur. For this reason, this embodiment does not target such an extremely rare case.

The present invention is not limited to the embodiment described above and illustrated in the drawings, and the following modifications or expansions are possible.
As the data structure of the packet P, the serial number S and the data part D may be interchanged.
The determination step of step A14 in FIG. 4 can be omitted when it is not necessary to assume a rare case where the same data as the header and footer exists in a part other than the header and footer in the packet. In this case, it may be configured to proceed to step A15 when “NO” in step A10.
Instead of step A3 in FIG. 4, a step of moving 1 byte from the reference address to the start address may be provided.
In the above embodiment, the example in which the present invention is applied to the controller 3 that controls the six-axis vertical articulated robot has been described. However, the present invention is a robot that controls the operation of the robot according to a control command given from the outside. It is applicable to all controllers.

  In the drawing, 2 is a robot, 3 is a controller (controller of the robot), 21 is a packet storage memory, and 22 is a control unit.

Claims (2)

A controller for a robot that controls the operation of the robot according to a control command given from the outside,
The control command is given in packet units having a predetermined data structure,
The packet includes a fixed value header provided at the beginning of the packet, a message length indicating the total data length of the packet provided immediately after the header, a serial number that is a unique value for each packet, It is composed of a data part indicating the control content and a fixed value footer provided at the end of the packet,
The header, the message length, the serial number and the footer are fixed length,
The data portion is of variable length;
A packet storage memory for temporarily storing the packet;
The received packets are sequentially stored from the start address to the end address of the packet storage memory in the order received, and it is determined whether or not the packet stored in the packet storage memory is normal. A control unit that controls the operation of the robot based on the control content indicated by the data part of the packet determined as
The controller is
A header search process for searching the header from a predetermined reference address toward the start address, and the total data length indicated by the message length provided immediately after the header when the header is found by the header search process And a footer in which the footer is stored using the header address in which the header found in the header search process is stored and the total data length acquired in the total data length acquisition process. A first confirmation process comprising a footer address detection process for detecting an address and a footer confirmation process for determining whether or not the footer exists in the footer address detected by the footer address detection process;
A second confirmation process for confirming whether or not the data length from the reference address to the head address of the header address matches the total data length acquired by the total data length acquisition process;
A normal data storage process for storing the serial number of the packet including the footer confirmed by the footer confirmation process;
Abnormal data indicating that the packet stored next to the end address side of the packet including the serial number is abnormal based on the serial number of the packet including the footer confirmed by the footer confirmation processing Abnormal data storage processing for storing
The stored serial number and the notification process for outputting the abnormal data to the outside are configured to be executable.
The controller is
When a packet is stored in the packet storage memory, the first confirmation process is executed with the last address of the packet as the reference address,
In the first confirmation process, when the header search process exceeds the start address, the notification process is executed, and when the footer confirmation process determines that the footer does not exist, the found header The header search process is performed from the header address toward the start address, and when the footer confirmation process determines that the footer exists, the second confirmation process is performed.
In the second confirmation process, when it is determined that the data length and the total data length match, the normal data storage process is executed, and it is determined that the data length and the total data length do not match. If the error data storage process is executed,
When the normal data storage process is executed, an address that is advanced by one address from the start address of the header of the packet including the stored serial number toward the start address side is used as the reference address. Execute the confirmation process
A controller for a robot, wherein the notification process is executed when the abnormal data storage process is executed.   In the second confirmation process, when it is determined that the data length and the total data length do not match, the control unit determines that the abnormal data if the data length is greater than the total data length. A storage process is executed, and when the data length is smaller than the total data length, the header search process is executed from the header address of the found header toward the start address. Item 2. The robot controller according to item 1.

Images