CN106625674B - An instruction processing method for a robot and the robot - Google Patents

Abstract

The invention discloses an instruction processing method for a robot and the robot. The method includes: controlling the host computer to receive and analyze the multimodal input data; controlling the host computer to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the analyzed multimodal input data; The execution instructions of the lower computer are stored in the instruction set; the upper computer is controlled to call out the execution instructions of the lower computers one by one from the instruction set and output to the lower computer; the lower computer is controlled to execute the execution instructions of the lower computer for multi-modal output. According to the method of the present invention, it can ensure that the execution instructions of the lower computer generated by the upper computer of the robot are sent to the lower computer for execution in an orderly manner, thereby greatly improving the operation reliability of the robot, ensuring the correctness of the output of the robot, and maintaining the user experience of the robot.

Description

An instruction processing method for a robot and the robot

technical field

The invention relates to the field of robots, in particular to an instruction processing method for a robot and the robot.

Background technique

With the continuous development of robot technology, more and more intelligent robots are applied to the daily production and life of human beings. With the continuous upgrading of intelligent robot hardware equipment, the functions of robots are becoming more and more complex, and the amount of data processing corresponding to robot operation is also increasing. The most direct manifestation is that during the operation of the robot, the amount of instructions generated and transmitted per unit time within the robot system continues to increase. In particular, in a robot system with an upper computer and a lower computer structure, during the interaction process between the upper computer and the lower computer, the lower computer instructions that the upper computer needs to send to the lower computer for execution per unit time continue to increase.

With the continuous upgrading of the processing capability of the robot system, in the implementation process of some more complex functional applications, the robot is faced with many execution actions, and it is difficult to implement each action in an orderly manner, which eventually leads to the robot Application functions cannot be perfectly realized, which greatly affects the user experience.

Contents of the invention

The invention provides an instruction processing method for a robot, the method comprising:

Control the host computer to receive and analyze multi-modal input data;

Controlling the upper computer to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the analyzed multimodal input data, and saving the lower computer execution instruction into the instruction set;

Control the upper computer to call out the lower computer to execute the instructions one by one from the instruction set and output them to the lower computer;

controlling the lower computer to execute the instructions executed by the lower computer to perform multimodal output.

In one embodiment, the method also includes:

Control the lower computer to judge the correctness of the instruction after receiving each execution instruction of the lower computer;

Control the lower computer to send the instruction to the upper computer after confirming that the received lower computer executes the instruction correctly;

Control the upper computer to delete each lower computer execution instruction from the instruction set until it receives the correct feedback of the instruction, and then continue to call out a lower computer execution instruction from the instruction set The command is output to the lower computer.

In one embodiment, control the host computer to call out the instructions from the instruction set to the lower computer to execute instructions one by one and output them to the lower computer, wherein an independent instruction calling thread is started to execute the operation of calling instructions from the instruction set one by one;

When the instruction set is empty, the instruction calling thread enters a sleep state;

The instruction calling thread is woken up after a preset specific period of time after the instruction calling thread enters the sleep state.

In one embodiment, the control host computer calls out the instructions from the instruction set to the lower computers to execute instructions one by one and outputs them to the lower computer, wherein:

The instructions in the instruction set are output to the lower computer in the order of first in first out.

In one embodiment, the communication between the upper computer and the lower computer is performed based on the Modbus protocol.

The present invention also proposes a robot, the robot includes a host computer and a lower computer, and the robot also includes:

The input data analysis module constructed in the host computer is configured to receive and analyze multimodal input data;

The lower computer instruction generation module constructed in the upper computer is configured to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the analyzed multimodal input data;

An instruction storage module constructed in the upper computer, configured to save the instructions executed by the lower computer;

The instruction output module constructed in the upper computer is configured to call out the lower computer execution instructions one by one from the instruction storage module and output them to the lower computer;

The instruction execution mechanism constructed in the lower computer is configured to execute the instruction executed by the lower computer to perform multimodal output.

In one embodiment:

The instruction execution mechanism includes an instruction verification unit, which is configured to judge the correctness of the instruction after receiving each execution instruction of the lower computer, and send the correct feedback of the instruction to the upper computer after determining that the execution instruction of the lower computer is correct. ;

The instruction output module is configured to delete the execution instruction of the lower computer from the instruction storage module and continue to save the instruction from the instruction storage module until the instruction is correctly fed back after sending each lower computer execution instruction. Call out a lower computer execution command from the module and output it to the lower computer.

In an embodiment, the instruction output module is configured to start an independent instruction calling thread to execute the operation of calling instructions one by one from the instruction storage module, wherein:

When the instruction storage module is empty, the instruction calling thread enters a sleep state;

The instruction calling thread is woken up after a preset specific period of time after the instruction calling thread enters the sleep state.

In an embodiment, the instruction output module is configured to output the instructions in the instruction storage module to the lower computer in a first-in first-out order.

In one embodiment, the communication between the upper computer and the lower computer is performed based on the Modbus protocol.

According to the method of the present invention, it can ensure that the execution instructions of the lower computer generated by the upper computer of the robot are sent to the lower computer for execution in an orderly manner, thereby greatly improving the operation reliability of the robot, ensuring the correctness of the output of the robot, and maintaining the user experience of the robot.

Additional features or advantages of the invention will be set forth in the ensuing description. And, some features or advantages of the present invention will be apparent from the description, or be understood by practicing the present invention. The objects and some of the advantages of the invention will be realized or obtained by the steps particularly pointed out in the written description, claims as well as the appended drawings.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the attached picture:

Fig. 1 is a method flowchart according to an embodiment of the present invention;

2 to 5 are partial flowcharts of methods according to embodiments of the present invention;

6 and 7 are schematic structural diagrams of a robot system according to an embodiment of the present invention.

Detailed ways

The implementation of the present invention will be described in detail below in conjunction with the accompanying drawings and examples, so that implementers of the present invention can fully understand how the present invention uses technical means to solve technical problems, and achieve the realization process of technical effects and according to the above-mentioned realization process The present invention is implemented concretely. It should be noted that, as long as there is no conflict, each embodiment and each feature in each embodiment of the present invention can be combined with each other, and the formed technical solutions are all within the protection scope of the present invention.

With the continuous upgrading of the processing capability of the robot system, in the implementation process of some more complex functional applications, the robot is faced with many execution actions, and it is difficult to implement each action in an orderly manner, which eventually leads to the robot Application functions cannot be perfectly realized, which greatly affects the user experience.

In view of the above problems, the present invention proposes an instruction processing method for a robot. In one embodiment of the present invention, after the upper computer of the robot generates an instruction that needs to be executed by the lower computer, it does not immediately send the instruction to the lower computer for execution, but saves the instruction in the instruction set, and then puts the instruction in the instruction set. The instructions are sent to the lower computer in sequence. In this way, when there are multiple lower computer instructions that need to be executed at the same time, there will be no instruction communication congestion and lower computer instruction execution confusion caused by simultaneous sending of instructions. In this way, the perfect execution of the output of the lower computer of the robot is guaranteed, and the user experience of the robot is greatly improved.

Specifically, in an embodiment of the present invention, the upper and lower computers of the robot system communicate based on the bus protocol (Modbus protocol) used in the industrial field.

Next, the detailed process of the method according to the embodiment of the present invention will be described in detail based on the accompanying drawings, and the steps shown in the flowchart of the accompanying drawings can be executed in a computer system including, for example, a set of computer-executable instructions. Although a logical order of steps is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

As shown in Figure 1, in one embodiment, first the robot controls the upper computer to receive multimodal input data (step S100); then controls the upper computer to analyze the received multimodal input data (step S110); then controls the upper computer Based on the multimodal input data after the analysis, generate the corresponding lower computer execution instruction (step S120) of the multimodal output data to be output, and save the lower computer execution instruction generated into the instruction set (step S130); next, the control The upper computer calls out the lower computer execution instructions one by one from the instruction set and outputs them to the lower computer (step S140); finally controls the lower computer to execute the lower computer execution instructions for multi-modal output.

Taking a specific application scenario as an example, assume that the robot needs the lower computer to execute commands A, B, and C in order to respond to the current interactive input data. In the prior art, when the lower computer executes instructions A, B, and C simultaneously (or sequentially in a short period of time) and simultaneously (or sequentially) output to the lower computer, it will inevitably cause communication congestion ( Generally, the communication channel between the upper and lower computers only supports the transmission of one instruction at the same time), and causes confusion in the execution of the lower computer (the lower computer receives multiple instructions to be executed at the same time, and it cannot confirm which one should be executed first). And in one embodiment of the present invention, when the upper computer generates the execution instructions A, B, and C of the lower computer, it is stored in the instruction set (A, B, C), and then the instructions in the instruction set are sequentially called out and output to The lower computer, that is to say, the lower computer receives instructions A, B, and C in sequence (one instruction at a time). In this way, not only the congestion of instruction transmission is avoided, but also the execution confusion of the lower computer instructions is avoided.

According to the method of the present invention, it can ensure that the execution instructions of the lower computer generated by the upper computer are sent to the lower computer for execution in an orderly manner, thereby greatly improving the operation reliability of the robot, ensuring the correctness of the output of the robot, and maintaining the user experience of the robot.

Further, in step S140, the upper computer calls out the lower computer execution instructions from the instruction set one by one and outputs them to the lower computer. Specifically, in one embodiment, the instructions in the instruction set are output to the lower computer in the order of first-in first-out. For example, if the upper computer generates the execution instructions A, B, and C of the lower computer in the order of A, B, and C, then after saving the execution instructions A, B, and C of the lower computer to the instruction set, the order of A, B, and C Invoke and output in turn. If the upper computer generates the execution instructions A, C, and B of the lower computer in the order of A, C, and B, then save the execution instructions A, C, and B of the lower computer to the instruction set, and then call them in the order of A, C, and B. and output.

In the embodiment shown in Figure 1, in step S140, the upper computer calls out the execution instructions of the lower computer from the instruction set one by one and outputs them to the lower computer. During this process, an error occurs in the output of a certain lower computer execution instruction Next, if you continue to output the next execution command of the lower computer in the original order, it will inevitably affect the overall multi-modal output of the lower computer. For example, the normal instruction sending execution sequence is A, B, C. If there is an error in the output of instruction B, then the instructions received by the lower computer are actually A, B1, C (B1 is an error instruction). In this way, not only the lack of output to be executed (less instruction B is executed), but also the execution confusion of the lower computer instructions (the failure to execute the wrong instruction B1 or the confusion of execution results will certainly disrupt the normal output results).

In view of the above problems, in one embodiment, a method of verification feedback of lower computer instructions is adopted. Specifically, after receiving each execution instruction of the lower computer, the control lower computer will judge the correctness of the instruction; after the control lower computer confirms that the execution instruction received by the lower computer is correct, it will send the correct instruction to the upper computer; After sending each lower computer execution command, until the correct feedback of the command is received, then delete the lower computer execution command from the instruction set, and continue to call out a lower computer execution command from the instruction set to output to the lower position machine.

Specifically, as shown in Figure 2, in one embodiment, the upper computer calls out the first lower computer execution instruction from the instruction set (the lower computer execution instruction with the highest output order) and outputs it to the lower computer (step S200) ; After receiving the instruction, the lower computer first verifies the instruction (step S210), and judges whether the instruction is correct (step S220).

If the lower computer judges that the execution instruction received by the lower computer is correct, then it sends the instruction and feeds it back to the upper computer (step S230). After the upper computer receives the correct feedback of the instruction, delete the first lower computer execution instruction (the output lower computer execution instruction) from the instruction set (step S240); then return to step S200, and call the current first lower computer execution instruction ( The lower computer with the highest current output order executes the instruction) and outputs to the lower computer.

If the lower computer judges that the execution instruction received by the lower computer is wrong, it sends an instruction error feedback to the upper computer (step S250). After receiving the instruction error feedback, the upper computer directly returns to step S200 (skip step S240), calls the first lower computer to execute the instruction and outputs it to the lower computer (outputs the output instruction again).

In this way, when the command output is wrong, the wrong command can be output repeatedly, so as to ensure that all the commands are correctly output to the lower computer in order.

Furthermore, in some instruction transmission scenarios, the result of the output error of the instruction is not only that the lower computer receives the wrong instruction, but also that the instruction is not sent to the lower computer at all. For example, the normal instruction sending execution sequence is A, B, C. If there is an error in the output of instruction B, then the instruction received by the lower computer is actually A, C (B1 is an error instruction). In this way, it not only causes the absence of the output to be executed (less instruction B is executed), but also causes confusion in the execution of the instructions of the lower computer (the absence of instruction execution will inevitably disrupt the normal order of instruction execution).

In view of this situation, in one embodiment of the present invention, after the upper computer outputs the command, it will judge whether the instruction given back by the lower computer within the preset period of time is correct. If there is no feedback, it means that the lower computer has not received the instruction (regardless of right or wrong). In this case, it is necessary to redo the command output.

Specifically, as shown in FIG. 3 , in one embodiment, the upper computer of the robot outputs the first execution instruction of the lower computer in the instruction set to the lower computer (step S300 ), and then starts timing (step S301 ).

During the timing process, it is judged whether the feedback result of whether the output command fed back by the lower computer is correct or not is received (step S310). If it is not received, then judge whether the current timing result has reached the preset duration, if not, continue timing and return to step S310, and continue to judge whether the feedback result of whether the output command fed back by the lower computer is received is correct . If the current timing result reaches the preset duration (the feedback from the lower computer is not received within the preset duration after the command is output), it means that the lower computer has not received the instruction sent by the upper computer, and returns to step S300 at this time, calling The first lower computer executes the instruction and outputs it to the lower computer (outputs the output instruction again).

In step S310, if the upper computer has received the feedback result of whether the output command fed back by the lower computer is correct, then stop timing (step S311), and whether the further instruction is correctly output (whether the lower computer feedbacks the instruction is correct or the instruction is wrong) ( Step S320). If the instruction output is correct, then delete the first lower-position computer execution instruction (the output lower-position computer execution instruction) (step S321) from the instruction set; then return to step S300, call the current first lower-position computer execution instruction (current output The lower computer with the highest order executes the instruction) and outputs to the lower computer. If the instruction output is wrong, then directly return to step S300, call the first lower computer to execute the instruction and output it to the lower computer (output the output instruction again).

Furthermore, in some instruction execution scenarios, the speed at which the lower computer executes instructions is much slower than the speed at which the upper computer generates and outputs new instructions. This will cause a large number of unexecuted lower computer execution instructions to accumulate at the lower computer, thereby causing the accumulation of instructions at the lower computer and affecting the smooth progress of the lower computer instructions.

In view of the above situation, in one embodiment, before sending the execution instruction of the lower computer, the upper computer first judges the current state of the lower computer, and only when the state of the lower computer is idle, the first lower computer execution instruction in the instruction set will be sent. to the next computer.

As shown in Figure 4, in one embodiment, the upper computer of the robot outputs the first lower computer execution instruction in the instruction set to the lower computer (step S400), and then judges whether the current instruction output is successful (correct) (step S410 ). Specifically, in step S410, it is judged whether the lower computer has received the instruction (whether there is feedback from the lower computer within the preset time length) and whether the instruction received by the lower computer is correct (whether the information fed back by the lower computer is correct or wrong) . When the upper computer judges that the current instruction output is wrong, it directly returns to step S300, calls the first lower computer to execute the instruction and outputs it to the lower computer (outputs the output instruction again).

When the upper computer judges that the current instruction output is correct, delete the first lower computer execution instruction (the output lower computer execution instruction) (step S420) from the instruction set; then judge whether the lower computer is in an idle state (whether it can receive new The lower computer executes the instruction) (step S430). If the lower computer is in an idle state, return to step S400, call the current first lower computer execution instruction (the lower computer execution instruction with the highest current output order) and output it to the lower computer. If the lower computer is not in the idle state, then wait (specifically, in this embodiment, the upper computer waits for a preset fixed duration), and return to step S430, and continue to judge whether the lower computer is in the idle state again.

Further, in order to ensure that the command output of the host computer can proceed smoothly, in an embodiment of the present invention, the host computer starts an independent command calling thread to execute operations of calling out commands one by one from the command set. In this way, the process of generating the command and the process of sending the command are separated from each other. The command generation process will not interfere with the command transmission, and a large number of commands will be generated at the same time (or in a short period of time), and the accumulation of command transmission and command communication congestion will not occur.

Specifically, as shown in Figure 5, in one embodiment, the upper computer first judges whether the instruction set storing the execution instructions of the lower computer is empty (step S500). The lower computer executes the instruction, and at this time, the application instruction calling thread invokes the first lower computer in the instruction set to execute the instruction and send it to the lower computer (step S550).

If the instruction set is empty, it means that there is currently no lower computer execution instruction that needs to be sent to the lower computer. At this time, the instruction calling thread enters a sleep state to reduce system resource consumption (step S510).

In order to prevent the instruction calling thread from being in the sleep state for a long time and missing the sending of the instruction, in one embodiment, the instruction calling thread is woken up after a preset specific period of time after the instruction calling thread enters the sleep state. Specifically, start timing when the instruction calling thread enters the sleep state (step S520), and judge whether the timing reaches a preset duration (step S530). If the preset duration is not reached, the sleep state of the instruction calling thread is continued, and timing is continued. If the preset duration is reached, wake up the instruction calling thread (step S540), return to step S500, and judge whether the current instruction set is empty.

To sum up, according to the method of the present invention, it can ensure that the execution instructions of the lower computer generated by the upper computer are sent to the lower computer for execution in an orderly manner, avoiding the occurrence of instruction accumulation, instruction error, and instruction loss. Thereby greatly improving the correct rate of command execution of the robot's lower computer, improving the reliability of the robot's operation, ensuring the correctness of the robot's output, and maintaining the user experience of the robot.

Based on the method of the invention, the invention also provides a robot. As shown in Figure 6, in one embodiment, the robot includes a host computer 601 and a lower computer 602, and the robot also includes:

The input data analysis module 610 constructed in the host computer 601 is configured to receive and analyze multimodal input data;

The lower computer instruction generation module 620 constructed in the upper computer 601 is configured to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the multimodal input data parsed by the input data analysis module 610;

The instruction storage module 630 constructed in the upper computer 601 is configured to save the lower computer execution instructions generated by the lower computer instruction generation module 620;

The instruction output module 640 constructed in the upper computer 601 is configured to call out the execution instructions of the lower computer from the instruction storage module 630 one by one and output them to the lower computer 602;

The instruction execution mechanism built in the lower computer 602 is configured to execute the received lower computer execution instruction to perform multi-modal output.

Specifically, in an embodiment, the instruction output module 640 is configured to output the instructions in the instruction storage module 630 to the lower computer 602 in a first-in first-out order.

Further, in an embodiment, the communication between the upper computer 601 and the lower computer 602 is performed based on the Modbus protocol.

Further, in one embodiment, the instruction execution mechanism on the lower computer includes an instruction verification unit, which is configured to judge the correctness of the instruction after receiving each lower computer execution instruction, and determine that the received lower computer execution instruction is correct Afterwards, the instruction is sent and fed back to the upper computer correctly; correspondingly, the instruction output module on the upper computer is configured to send each lower computer execution instruction until it receives correct feedback of the instruction, and then sends the lower computer execution instruction from the Delete from the instruction storage module, and continue to call out a lower computer execution instruction from the instruction storage module to output to the lower computer.

As shown in FIG. 7 , in an embodiment, the upper computer 701 is configured with an input data analysis module 710 , a lower computer instruction generation module 720 , an instruction storage module 730 and an instruction output module 740 .

The input data analysis module 710 receives and analyzes the multimodal input data; the lower computer instruction generation module 720 generates a lower computer execution instruction corresponding to the multimodal output data to be output based on the multimodal input data analyzed by the input data analysis module 710; The instruction saving module 730 saves the execution instructions of the lower computer generated by the lower computer instruction generation module 720 ;

The lower computer 702 is configured with an instruction execution mechanism 750, and the instruction verification unit 751 judges the correctness of the instruction after the instruction execution mechanism 750 receives each lower computer execution instruction, and sends the correct instruction after confirming that the received lower computer execution instruction is correct. Feedback to the instruction output module 740 of the host computer 701. Correspondingly, after the instruction output module 740 sends each lower computer execution instruction, it will delete the lower computer execution instruction from the instruction storage module 730 until it receives the correct feedback from the instruction verification unit 751, and continue to save the instruction from the instruction storage module 730. In the module 730, a lower computer execution instruction is called out to the lower computer 702.

Further, in an embodiment, the instruction output module is configured to start an independent instruction calling thread to execute the operation of calling instructions one by one from the instruction storage module, wherein:

When the instruction saving module is empty, the instruction calling thread enters the sleep state;

When the instruction calling thread enters the sleep state, the instruction calling thread is woken up after a preset specific time period.

Although the embodiments disclosed in the present invention are as above, the described content is only an embodiment adopted for the convenience of understanding the present invention, and is not intended to limit the present invention. The method described in the present invention can also have other various embodiments. Without departing from the essence of the present invention, those skilled in the art may make various corresponding changes or modifications according to the present invention, but these corresponding changes or modifications shall belong to the protection scope of the claims of the present invention.

Claims (8)

1. A command processing method for a robot, characterized in that the method comprises: Control the host computer to receive and analyze multi-modal input data; Controlling the upper computer to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the analyzed multimodal input data, and saving the lower computer execution instruction into the instruction set; Controlling the upper computer to call out the execution instructions of the lower computers one by one from the instruction set and output them to the lower computer, including: controlling the lower computer to judge the correctness of the instructions after receiving each lower computer execution instruction; After controlling the lower computer to determine whether the execution command received by the lower computer is correct, send the feedback result of whether the command is correct to the upper computer; After the control host computer sends each execution instruction of the lower computer, it judges whether the instruction is received within the preset time period and whether the feedback result is correct or not, wherein, If the result of the correct feedback of the instruction is received within the preset time length, the outputted lower computer execution instruction is deleted from the instruction set, and it is further judged whether the lower computer is in an idle state. If it is in an idle state, then From the instruction set, the current first lower computer execution instruction is transferred to the lower computer, if not in an idle state, then wait and continue to judge whether the lower computer is in an idle state; If the result of the command error feedback is received within the preset time period, or the result of whether the command is correct or not is not received within the preset time period, the lower computer with the highest current output order is called to execute the command to drive The lower computer will output the output command again; controlling the lower computer to execute the instructions executed by the lower computer to perform multimodal output. 2. The method according to claim 1, characterized in that, controlling the upper computer to call out the execution instructions of the lower computer from the instruction set one by one and output them to the lower computer, wherein starting an independent instruction call thread to execute instructions from the instruction set Call out the operations of instructions one by one; When the instruction set is empty, the instruction calling thread enters a sleep state; The instruction calling thread is woken up after a preset specific period of time after the instruction calling thread enters the sleep state. 3. The method according to claim 1, wherein the control host computer calls out the lower computer to execute instructions one by one from the instruction set and outputs them to the lower computer, wherein: The instructions in the instruction set are output to the lower computer in the order of first in first out. 4. The method according to claim 1, wherein the communication between the upper computer and the lower computer is performed based on the Modbus protocol. 5. A robot, characterized in that, the robot includes a host computer and a lower computer, and the robot also includes: The input data analysis module constructed in the host computer is configured to receive and analyze multimodal input data; The lower computer instruction generation module constructed in the upper computer is configured to generate a lower computer execution instruction corresponding to the multimodal output data to be output based on the analyzed multimodal input data; An instruction storage module constructed in the upper computer, configured to save the instructions executed by the lower computer; The instruction output module constructed in the upper computer is configured to call out the lower computer execution instructions one by one from the instruction storage module and output them to the lower computer; The instruction execution mechanism constructed in the lower computer is configured to execute the execution instructions of the lower computer to perform multimodal output, wherein the robot further includes: the instruction execution mechanism includes an instruction verification unit, which is configured to After receiving each execution instruction of the lower computer, judge the correctness of the instruction, and after determining whether the execution instruction of the lower computer received is correct, send the feedback result of whether the instruction is correct to the upper computer; The instruction output module is configured to determine whether the instruction is received within a preset time period and whether the instruction is correct or not after sending each lower computer execution instruction, wherein, If the result of the correct feedback of the instruction is received within the preset time length, the outputted lower computer execution instruction will be deleted from the instruction storage module, and it will be further judged whether the lower computer is in an idle state, if it is in an idle state , then transfer the current first lower computer execution instruction from the instruction storage module to output to the lower computer, if not in an idle state, then wait and continue to judge whether the lower computer is in an idle state; If the result of the command error feedback is received within the preset time period, or the result of whether the command is correct or not is not received within the preset time period, the lower computer with the highest current output order is called to execute the command to drive The lower computer outputs the command that has been outputted again. 6. The robot according to claim 5, wherein the instruction output module is configured to start an independent instruction calling thread to perform the operation of calling out instructions one by one from the instruction storage module, wherein: When the instruction storage module is empty, the instruction calling thread enters a sleep state; The instruction calling thread is woken up after a preset specific period of time after the instruction calling thread enters the sleep state. 7. The robot according to claim 5, wherein the instruction output module is configured to output the instructions in the instruction storage module to the lower computer in a first-in first-out order. 8. The robot according to claim 5, wherein the communication between the upper computer and the lower computer is performed based on the Modbus protocol.