Control method of yeast treading robot, electronic equipment and computer program product

Abstract

According to the control method of the yeast treading robot, the electronic equipment and the computer program product, the surface of a yeast mold is divided into a plurality of grids, independent detection and treading control of each area are achieved, uniformity of parameters (such as height and density) of all parts of a yeast block is ensured, and the problem that the tightness degree of manual yeast treading is not consistent is solved. In the yeast treading process, based on real-time calculation of the target parameter matching degree (such as height and density), only the grids which do not reach the standard are treaded in a targeted mode, invalid repeated actions are avoided, and efficiency is improved. According to the embodiment, the sequence is repeatedly treaded until parameters reach the standard, the manual repeated treading process is simulated, it is ensured that the curved block reaches the ideal structure with the tight outer portion and the loose inner portion, meanwhile, surface slurry forming is promoted, and the slurry extraction effect is improved. Moreover, according to the yeast treading robot, full-automatic yeast treading is achieved, and the labor intensity and the labor cost are remarkably reduced.

Description

Technical Field

[0001] This application relates to the field of robotics technology, and more specifically, to a control method, electronic equipment, and computer program product for a treadle robot. Background Technology

[0002] Yeast starter is an indispensable raw material in the brewing process. It is typically made from barley, wheat, peas, etc., which are crushed, mixed with water, kneaded, and pressed into brick-shaped blocks (called block starter or brick starter). These blocks then undergo natural fermentation by microorganisms. In the brewing process, yeast starter is added after the grain mash is steamed to promote subsequent fermentation. Currently, yeast starter preparation mainly relies on two methods: manual yeast making and mechanized yeast making.

[0003] Artificial koji making employs the traditional method of manual treading, where materials are placed in molds and workers repeatedly tread on them barefoot to shape them. The advantages of this method are: through repeated treading, a layer of slurry forms on the surface of the koji blocks, acting as a seal to prevent the invasion of harmful microorganisms and the loss of internal nutrients, thus ensuring fermentation quality; simultaneously, manual treading creates a structure that is tight on the outside and loose on the inside, which is beneficial for fermentation. However, artificial koji making has the following problems: it is difficult to maintain consistent treading force and tightness, resulting in inconsistent koji block quality; and it is labor-intensive and has high labor costs.

[0004] Mechanized koji making uses koji-making machines to press koji blocks into shape in one or more directions, significantly improving production efficiency and reducing labor intensity. However, mechanized koji making lacks the process of repeated manual trampling, resulting in poor surface slurry extraction from the koji blocks, affecting sealing and fermentation quality, and consequently reducing the quality of subsequent brewing. Summary of the Invention

[0005] The purpose of this application is to provide a control method, electronic device and computer program product for a treading robot, so as to solve the problem that the prior art is difficult to balance pulp quality, production efficiency and labor cost.

[0006] This application provides a control method for a treadmill robot, applied to the controller of the treadmill robot, the method comprising: The surface of the curved mold is divided into multiple grids; For each grid cell, the target parameter matching degree is calculated. If the target parameter matching degree is less than the matching degree threshold, the corresponding grid cell is added to the trampling sequence. Perform one round of trampling on all grids in the trampling sequence; Repeat the above process until the target parameter matching degree of all grids is greater than or equal to the matching degree threshold, or the number of pedaling cycles exceeds the set threshold, then end the pedaling process.

[0007] In the above technical solution, by dividing the surface of the curved mold into multiple grids, independent detection and treading control of each area are achieved, ensuring the uniformity of parameters (such as height and density) of each part of the curved block and avoiding the problem of inconsistent tightness caused by manual treading. During the treading process, based on the real-time calculation of the target parameter matching degree (such as height and density), only the grids that do not meet the standard are targeted for treading, avoiding ineffective repetitive actions and improving efficiency. This embodiment simulates the manual repetitive treading process by repeating the treading sequence until the parameters meet the standard, ensuring that the curved block achieves the ideal structure of being tight on the outside and loose on the inside, while promoting the formation of surface slurry and improving the slurry lifting effect. Furthermore, the treading robot in this embodiment achieves fully automated treading, significantly reducing labor intensity and labor costs.

[0008] In some alternative implementations, the target parameter matching degree includes a high degree of matching; The method for calculating the high degree of matching includes: Use a depth camera to obtain a height map, and get the height h(x, y) of the grid. The high degree of matching is: ; Among them, h * (x, y) represents the optimal block height, where x and y are the grid plane coordinates.

[0009] The above technical solution introduces a formula for calculating the height matching degree, compares the actual height with the optimal block height, and uses mathematical quantification to determine whether stepping is necessary, thereby improving process consistency.

[0010] In some optional implementations, the target parameter matching degree includes density matching degree; The method for calculating the density matching degree includes: The force spectrum is recorded during the treading process; where the force spectrum is the change of the applied force over time during the treading process. The force spectrum matching degree is: ; Where F(x, y, t) is the force spectrum; F * (x, y, t) represents the empirically optimal force spectrum, t represents the current time, and T represents the time taken for a single pedal stroke; The force spectrum matching degree is used as the density matching degree.

[0011] In the above technical solution, by recording the force spectrum data during the treading process in real time, the dynamic changes in the internal density of the block are directly reflected. The actual force spectrum is compared with the preset empirical optimal force spectrum, and the need for treading is determined through mathematical quantification, thereby improving the consistency of the process.

[0012] In some alternative implementations, performing one round of stomping on all grids in the stomping sequence includes: Based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model, force-position hybrid control of pedaling is achieved.

[0013] The above technical solution adopts force-position mixing control technology, which transforms the experience-based operation of traditional koji making process into a quantifiable and optimizable automated control process. It retains the core advantages of manual koji making while solving the problems of insufficient precision and low efficiency of mechanized koji making, and realizes a highly efficient, stable and replicable koji making process.

[0014] In some optional implementations, the location cost function is: ; The speed cost function is: ; Where q is the joint space coordinate, q d Let x be the target space coordinates of the joint. i Let x be the task space coordinate. i d J represents the spatial coordinates of the mission target. i For the task Jacobian matrix, W i As the task weight, W q Here, N represents the joint weight, and N represents the total number of tasks.

[0015] In the above technical solution, the position cost function ensures that the robot's foot is accurately positioned to the target grid by minimizing the deviation between the joint space coordinates and the target coordinates, as well as the deviation between the task space coordinates and the target coordinates. The velocity cost function, combining the Jacobian matrix and task weights, optimizes the matching degree between joint velocity and task space velocity.

[0016] In some alternative implementations, the acceleration cost function is: ; The robot's dynamics model is as follows: ; ; ; Where, f c For the foot end to bear the force, f d For the target at the foot, J c Let τ be the Jacobian matrix of the forces acting on the foot. d For the target torque of the joint, W f H is the force weight at the foot, C is the inertia matrix, G is the centrifugal force matrix, and W is the gravity term. τFor joint torque weighting, The range of force applied to the foot. The target torque of the joint is the minimum value. This represents the maximum target torque of the joint.

[0017] In the above technical solution, the acceleration cost function reduces mechanical vibration and inertial impact by optimizing the deviation between joint acceleration and target acceleration. The robot dynamics model incorporates an inertia matrix, a centrifugal force matrix, and a gravity term to compensate for the robot's own dynamic effects in real time.

[0018] In some optional implementations, the stepping control based on position cost function, velocity cost function, acceleration cost function, and robot dynamics model to achieve force-position hybrid control includes: Non-contact phase: For the stepped grid (x, y), control the foot to move to coordinate (x, y, h), and simultaneously apply force f to the foot. c The force in the x, y, and z directions is limited to 0 until the foot touches the curved block; where h is the height of the foot. In the non-contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

[0019] In the above technical solution, during the non-contact phase, the foot is controlled to move to the target coordinates while the forces on the X / Y / Z axes are limited to zero, ensuring that the foot does not deviate or collide before contacting the curved block. During this process, the target joint coordinates, velocity, and torque are calculated in real time using position, velocity, acceleration cost functions, and dynamic models, and then converted into motor commands to achieve smooth and rapid motion control.

[0020] In some optional implementations, the stepping control that achieves force-position hybrid control based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model further includes: Contact phase: After the foot contacts the curved block, the force f applied to the foot will be... c The force in the x and y directions is limited to 0, controlling the force f on the foot. c The force in the z-direction is: ; Among them, K F K h This is the proportionality coefficient; During the contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

[0021] In the above technical solution, during the contact phase, the force in the X / Y directions is limited to zero, and the pedaling force is dynamically adjusted only through the Z-direction force to simulate the manual bending process. During this process, the target coordinates, velocity, and torque of the joint are calculated in real time using position, velocity, acceleration cost functions, and dynamic models, and then converted into motor commands to achieve smooth and rapid motion control.

[0022] In some alternative implementations, multiple pressure detection units are provided at different positions on the bottom of the foot of the treading robot.

[0023] In some optional implementations, the stepping control that achieves force-position hybrid control based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model further includes: Determine whether the difference between the average pressures on both sides is greater than the first lateral deviation threshold and the duration is greater than the corresponding threshold. If so, continue to determine whether the difference between the average pressures on both sides is less than the second lateral deviation threshold. If the difference in average pressure between the two sides is less than the second lateral deviation threshold, then control the foot's fine-tuning angle: ; Where, k θ PL and PR are proportional parameters, and PL and PR are the average values ​​of the pressure on the opposite side. If the average pressure on both sides is greater than or equal to the second lateral deviation threshold, then reduce the force f on the foot. c Apply force in the z-direction and control the foot to return to a safe height, then replan the pedaling motion.

[0024] In the above technical solution, lateral deviation dynamic adjustment is achieved through multiple pressure detection units on the sole of the foot.

[0025] An electronic device provided in this application includes a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions, when executed by the processor, perform any of the methods described above.

[0026] This application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of any of the methods described above. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This application illustrates the possible treading system structure in its embodiments; Figure 2 A schematic diagram of the joint degrees of freedom of a treading robot provided in an embodiment of this application; Figure 3 A schematic diagram of the joint degrees of freedom of a treading robot provided in another embodiment of this application; Figure 4 This is a schematic diagram of the ankle drive method of a stepping robot provided in an embodiment of this application; Figure 5 This is a schematic diagram of the ankle-driven mechanism for a stepping robot provided in another embodiment of this application; Figure 6 A flowchart illustrating the steps of a control method for a treading robot provided in this application embodiment; Figure 7 A possible structural schematic diagram of an electronic device provided in an embodiment of this application; Figure 8 This is a schematic diagram of the treadmill production line provided in some embodiments of this application; Figure 9 This is a schematic diagram of the structure of the bending device provided in some embodiments of this application; Figure 10 This is a schematic diagram of the structure of the bending device provided in some embodiments of this application; Figure 11 This is a schematic diagram of the structure of the bending device provided in some embodiments of this application; Figure 12 for Figure 11 A magnified view of a portion of region A in the middle; Figure 13 This is an assembly diagram of the buffer plate and elastic buffer provided in some embodiments of this application; Figure 14 Exploded views of the foot assembly provided in some embodiments of this application; Figure 15 This is an assembly diagram of the lower leg assembly and foot assembly provided in some embodiments of this application; Figure 16This is a schematic diagram of the structure of the leg power module provided in some embodiments of this application; Figure 17 This is an assembly diagram of a joint drive unit provided in some embodiments of this application; Figure 18 This is an assembly diagram of a joint drive unit provided in some embodiments of this application. Detailed Implementation

[0029] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0030] Please refer to Figure 1 , Figure 1 This paper illustrates the structure of a koji-making system that may be employed in embodiments of this application. The koji-making system includes a koji-making robot, a depth camera, and a conveyor belt. The depth camera is used to acquire real-time height maps of the koji mold surface, obtaining height data for each grid cell, providing visual data support for calculating height matching accuracy. The conveyor belt transports the koji mold to the working area of ​​the koji-making robot, enabling continuous production. The koji mold carries brewing raw materials (such as a mixture of barley and wheat) and is pressed into koji blocks by the koji-making robot. The koji-making robot has a multi-degree-of-freedom robotic arm, supporting flexible movement to adapt to the koji-making needs of different koji mold areas; the robot's foot integrates multiple pressure detection units to monitor the koji pressure distribution in real time.

[0031] The flexing robot's joint motors have built-in position sensors; alternatively, it may use joint motors without position sensors, but install position sensors on each joint. The flexing robot may also use joint motors with force sensing capabilities; or, it may use joint motors without force sensing capabilities, but install force sensors on each joint.

[0032] The treadmill robot can be used as follows Figure 2 The joint degrees of freedom shown can also be represented as follows: Figure 3 The joint degrees of freedom are shown. The difference between the two types of joint degrees of freedom lies in the different combinations of the first, second, and third degrees of freedom. It should be noted that, in addition to the two types of joint degrees of freedom mentioned above, the control method of this scheme can also be applied to other combinations of joint degrees of freedom.

[0033] In a flexing robot, the motor can be directly mounted at the joint position, such as... Figure 4 As shown.

[0034] In a treadmill robot, the motor can also be installed in another location and the joints can be driven using linkages, belts, ropes, etc., for example... Figure 5 As shown, the fifth and sixth degrees of freedom at the ankle are driven by the drive linkage.

[0035] Please refer to Figure 6 , Figure 6 A flowchart illustrating the steps of a control method for a treadmill robot provided in this application embodiment, applied to the controller of the treadmill robot, the method comprising: Step S1: Divide the surface of the curved mold into multiple grids; Step S2: For each grid cell, calculate the target parameter matching degree. If the target parameter matching degree is less than the matching degree threshold, add the corresponding grid cell to the trampling sequence. Step S3: Perform one round of trampling on all grids in the trampling sequence; Step S4: Repeat steps S2 and S3 until the target parameter matching degree of all grids is greater than or equal to the matching degree threshold, or the number of pedaling cycles exceeds the set threshold, then end the pedaling process.

[0036] In this embodiment, the surface of the curved mold is divided into multiple grids, enabling independent detection and treading control of each area. This ensures the uniformity of parameters (such as height and density) across the curved block, avoiding the inconsistency in tightness caused by manual treading. During the treading process, based on real-time calculation of the target parameter matching degree (such as height and density), only grids that do not meet the standards are targeted for treading, avoiding ineffective repetitive actions and improving efficiency. This embodiment simulates the manual treading process by repeating the treading sequence until the parameters meet the standards, ensuring that the curved block achieves the ideal structure of being tight on the outside and loose on the inside, while promoting the formation of surface slurry and improving the slurry lifting effect. Furthermore, the treading robot in this embodiment achieves fully automated treading, significantly reducing labor intensity and labor costs.

[0037] In some alternative implementations, the target parameter matching degree includes a high degree of matching; The method for calculating the high degree of matching includes: Use a depth camera to obtain a height map, and get the height h(x, y) of the grid. The high degree of matching is: ; Among them, h * (x, y) represents the optimal block height, where x and y are the grid plane coordinates.

[0038] In this embodiment, a formula for calculating the height matching degree is introduced to compare the actual height with the optimal block height. The mathematical quantification determines whether stepping is necessary, thereby improving process consistency.

[0039] In some optional implementations, the target parameter matching degree includes density matching degree; The method for calculating the density matching degree includes: The force spectrum is recorded during the treading process; where the force spectrum is the change of the applied force over time during the treading process. The force spectrum matching degree is: ; Where F(x, y, t) is the force spectrum; F * (x, y, t) represents the empirically optimal force spectrum, t represents the current time, and T represents the time taken for a single pedal stroke; The force spectrum matching degree is used as the density matching degree.

[0040] In this embodiment of the application, by recording the force spectrum data during the treading process in real time, the dynamic changes in the internal density of the block are directly reflected. The actual force spectrum is compared with the preset empirical optimal force spectrum, and the need for treading is determined through mathematical quantification, thereby improving the consistency of the process.

[0041] The relationship between density and force spectrum is explained as follows: Assuming the pressure surface of the curved foot is approximately planar, the change in height Δh of the curved block and the pressure F satisfy the following: ; Where hi is the initial height, A is the contact area between the foot and the curved surface, and E is the material modulus of the curved block. Therefore, the density ρ can be approximated as: ; Where M is weight and V is volume.

[0042] It can be seen that within the applicable range, ρ increases monotonically with F and changes approximately linearly, so the force spectrum can be used to characterize the block density.

[0043] In step S2, the situation where the target parameter matching degree of the raster is lower than the set threshold can be divided into the following three cases: the height matching degree is lower than the corresponding threshold; the density matching degree is lower than the corresponding threshold; or either the height matching degree or the density matching degree is lower than the corresponding threshold.

[0044] The calculation of density matching degree relies on the force spectrum data recorded during the previous trampling process. If it is the first trampling (no historical force spectrum data), one of the following two methods can be used: trampling according to the preset initial force; or directly using the preset initial force feedback value as the calculation benchmark.

[0045] In some alternative implementations, performing one round of stomping on all grids in the stomping sequence includes: Based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model, force-position hybrid control of pedaling is achieved.

[0046] In this embodiment, force-position mixing control technology is used to transform the experience-based operation of traditional koji-making process into a quantifiable and optimizable automated control process. This retains the core advantages of manual koji-making while solving the problems of insufficient precision and low efficiency in mechanized koji-making, thus achieving a highly efficient, stable, and reproducible koji-making process.

[0047] In some optional implementations, the location cost function is: ; The speed cost function is: ; Where q is the joint space coordinate, q d Let x be the target space coordinates of the joint. i Let x be the task space coordinate. i d J represents the spatial coordinates of the mission target. i For the task Jacobian matrix, W i As the task weight, W q Here, N represents the joint weight, and N represents the total number of tasks.

[0048] In this embodiment, the position cost function ensures precise positioning of the robot's foot to the target grid by minimizing the deviation between joint space coordinates and target coordinates, as well as the deviation between task space coordinates and target coordinates. The velocity cost function, combining the Jacobian matrix and task weights, optimizes the matching degree between joint velocity and task space velocity.

[0049] In some alternative implementations, the acceleration cost function is: ; The robot's dynamics model is as follows: ; ; ; Where, f c For the foot end to bear the force, f d For the target at the foot, J c Let τ be the Jacobian matrix of the forces acting on the foot. d For the target torque of the joint, W f H is the force weight at the foot, C is the inertia matrix, G is the centrifugal force matrix, and W is the gravity term. τ For joint torque weighting, The range of force applied to the foot. The target torque of the joint is the minimum value. This represents the maximum target torque of the joint.

[0050] In this embodiment, the acceleration cost function reduces mechanical vibration and inertial impact by optimizing the deviation between joint acceleration and target acceleration. The robot dynamics model incorporates an inertia matrix, a centrifugal force matrix, and a gravity term to compensate for the robot's own dynamic effects in real time.

[0051] In some optional implementations, the stepping control based on position cost function, velocity cost function, acceleration cost function, and robot dynamics model to achieve force-position hybrid control includes: Non-contact phase: For the stepped grid (x, y), control the foot to move to coordinate (x, y, h), and simultaneously apply force f to the foot. c The force in the x, y, and z directions is limited to 0 until the foot touches the curved block; where h is the height of the foot. In the non-contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

[0052] In this embodiment, during the non-contact phase, the foot is controlled to move to the target coordinates while the forces on the X / Y / Z axes are limited to zero, ensuring that the foot does not deviate or collide before contacting the curved block. During this process, the target joint coordinates, velocity, and torque are calculated in real time using position, velocity, acceleration cost functions, and a dynamic model, and then converted into motor commands to achieve smooth and rapid motion control.

[0053] In some optional implementations, the stepping control that achieves force-position hybrid control based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model further includes: Contact phase: After the foot contacts the curved block, the force f applied to the foot will be... c The force in the x and y directions is limited to 0, controlling the force f on the foot. c The force in the z-direction is: ; The first term is the optimal force spectrum, the second term is the self-adjustment based on the force spectrum error, and the third term is the self-adjustment based on the height error. K F K h This is the proportionality coefficient; During the contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

[0054] In the above technical solution, during the contact phase, the force in the X / Y directions is limited to zero, and the pedaling force is dynamically adjusted only through the Z-direction force to simulate the manual bending process. During this process, the target coordinates, velocity, and torque of the joint are calculated in real time using position, velocity, acceleration cost functions, and dynamic models, and then converted into motor commands to achieve smooth and rapid motion control.

[0055] After receiving the motor position, speed, and current commands, the motor terminal executes these commands using a fast current loop (high frequency ~10kHz) to ensure accurate output. ; in, To form the final current command after integrating the motor position / speed / current commands, Kp / Kd are the proportional / derivative parameters.

[0056] During the trampling process, the weight W is used at different stages. i W f W τ W q The changes should be made accordingly, including: During the non-contact phase, the control objectives are primarily high positional accuracy and low force output, and the corresponding weights can be set as: W i For high, W f W is 0 τ Medium, W q It is of medium quality.

[0057] The contact phase can be further divided into the soft landing contact phase, the compaction phase, and the lifting phase. During the soft-landing contact phase, the control objectives are smooth contact and avoiding impact, and the corresponding weights can be set as: W i Medium, W f Medium, W τ Medium, W q It is of medium quality.

[0058] During the compaction stage, the control objective is to prioritize force spectrum tracking, and the corresponding weight can be set as: W i For low, W f For high, W τ W is on the high side. q It is of medium quality.

[0059] During the leg lift phase, the weight can be set to: W i For low, W f Medium, W τ W is on the high side. q It is of medium quality.

[0060] In some alternative implementations, multiple pressure detection units are provided at different positions on the bottom of the foot of the treading robot.

[0061] In some optional implementations, the stepping control that achieves force-position hybrid control based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model further includes: Determine whether the difference between the average pressures on both sides is greater than the first lateral deviation threshold and the duration is greater than the corresponding threshold. If so, continue to determine whether the difference between the average pressures on both sides is less than the second lateral deviation threshold. If the difference in average pressure between the two sides is less than the second lateral deviation threshold, then control the foot's fine-tuning angle: ; Where, k θ PL and PR are proportional parameters, and PL and PR are the average values ​​of the pressure on the opposite side. If the average pressure on both sides is greater than or equal to the second lateral deviation threshold (or a large lateral force occurs), then reduce the force f on the foot. c Apply force in the z-direction and control the foot to return to a safe height, then replan the pedaling motion.

[0062] In this embodiment, lateral deviation dynamic adjustment is achieved through multiple pressure detection units on the sole of the foot.

[0063] Figure 7 This illustration shows a possible structure of an electronic device provided in an embodiment of this application. (Refer to...) Figure 7 The electronic device includes a processor, memory, and a communication interface, which are interconnected and communicate with each other via a communication bus and / or other forms of connection mechanism (not shown).

[0064] The memory includes one or more (only one is shown in the figure), which can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The processor and other possible components can access the memory to read and / or write data to it.

[0065] The processor comprises one or more (only one is shown in the figure), which can be an integrated circuit chip with signal processing capabilities. The aforementioned processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Microcontroller Unit (MCU), a Network Processor (NP), or other conventional processors; it can also be a special-purpose processor, including a Neural-network Processing Unit (NPU), a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. Furthermore, when there are multiple processors, some can be general-purpose processors, and others can be special-purpose processors.

[0066] The communication interface includes one or more (only one is shown in the figure), which can be used to communicate directly or indirectly with other devices to exchange data. The communication interface may include interfaces for wired and / or wireless communication.

[0067] One or more computer program instructions may be stored in the memory, and the processor may read and execute these computer program instructions to implement the methods provided in the embodiments of this application.

[0068] Understandable. Figure 7 The structure shown is for illustrative purposes only; the electronic device may also include structures that are more complex than those shown. Figure 7 The more or fewer components shown, or having the same Figure 7 The different structures shown. Figure 7 The components shown can be implemented using hardware, software, or a combination thereof. Electronic devices may be physical devices, such as PCs, laptops, tablets, mobile phones, servers, embedded devices, etc., or they may be virtual devices, such as virtual machines, virtualized containers, etc. Furthermore, electronic devices are not limited to a single device; they can also be a combination of multiple devices or a cluster of a large number of devices.

[0069] This application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of any of the methods described above.

[0070] Please refer to Figure 8 , Figure 8 This is a schematic diagram of the structure of a koji-making production line 100 provided in some embodiments of this application. The koji-making production line 100 includes a conveyor belt 20 and a koji-making device 10. The conveyor belt 20 is used to transport the koji mold 30 containing koji 40, and the koji-making device 10 is used to tread on the koji 40 located in the koji mold 30.

[0071] One or more yeast molds 30 can be transported on the conveyor belt 20. For example, multiple yeast molds 30 containing yeast 40 can be transported on the conveyor belt 20, and the multiple yeast molds 30 can be transported to the yeast-pressing device 10 in sequence to complete the yeast-pressing process of the yeast 40.

[0072] The treading production line 100 may include only one treading device 10 or multiple treading devices 10. In an embodiment where the treading production line 100 includes multiple treading devices 10, the multiple treading devices 10 are arranged at intervals along the conveying direction of the conveyor belt 20.

[0073] Please refer to Figure 9 , Figure 9 This is a schematic diagram of the structure of a treading device 10 provided in some embodiments of this application. Embodiments of this application provide a treading device 10, including a frame 1, a robot 2, a vision detection unit 3, and a control unit 4. The robot 2 includes a torso body 21 and leg power modules 22. The torso body 21 is suspended from the frame 1, and the leg power modules 22 are connected to the waist of the torso body 21. The leg power modules 22 have multiple joint drive units 221. The vision detection unit 3 is mounted on the frame 1 and is used to acquire image information of the yeast 40. The control unit 4 is communicatively connected to the vision detection unit 3 and electrically connected to the multiple joint drive units 221. The control unit 4 is used to control the movement of the multiple joint drive units 221 according to the image information, so that the leg power modules 22 tread on the yeast 40.

[0074] The frame 1 can be a frame structure, and both the robot 2 and the vision inspection unit 3 are mounted on the frame 1. The torso 21 of the robot 2 can include a chest and a waist, with the chest connecting the waist and the frame 1. The torso 21 can be welded to the frame 1; the torso 21 can be snapped onto the frame 1; or the torso 21 can be connected to the frame 1 by fasteners, such as bolts or rivets. The leg power module 22 can be fixedly connected to the waist of the torso 21; for example, the leg power module 22 and the waist of the torso 21 can be connected by bolts; or the leg power module 22 can be movably connected to the waist of the torso 21; for example, the leg power module 22 and the torso 21 can be connected by a hip joint assembly, wherein the hip joint assembly includes a drive motor and a connecting block, the connecting block connects the power output end of the leg power module 22 and the drive motor, the drive motor is connected to the waist of the torso 21, and the drive motor can drive the leg power module 22 to rotate relative to the waist.

[0075] The leg power module 22 includes a plurality of joint drive units 221, which can drive multiple joint movements of the leg power module 22. In embodiments where the hip joint assembly includes a drive motor, the plurality of joint drive units 221 include a drive motor.

[0076] The treading device 10 has a treading station, and the leg power module 22 can tread the yeast 40 located at the treading station. The vision inspection unit 3 is mounted on the frame 1, and the vision inspection unit 3 can acquire image information of the yeast 40 at the treading station. The vision inspection unit 3 can be a camera, laser profilometer, etc.

[0077] The control unit 4 can be mounted on the frame 1; it can be mounted on the torso body 21; or it can be placed separately from the frame 1, with the control unit 4 electrically connected to the vision detection unit 3 and the joint drive unit 221 via wires. The control unit 4 can be a microcontroller, servo controller, etc. For example, such as... Figure 9 As shown, the chest has a thoracic cavity, and the control unit 4 is located inside the thoracic cavity and partially protrudes from the chest.

[0078] The control unit 4 can control the joint drive unit 221 to adjust the position and angle of the leg power module 22 stepping on the yeast 40 according to the image information.

[0079] In this embodiment, the control unit 4 can control the movement of multiple joint drive units 221 of the leg power module 22 to make the leg power module 22 tread on the yeast 40. The vision detection unit 3 can acquire image information of the shape of the yeast 40, and the control unit 4 can send instructions to the multiple joint drive units 221 of the leg power module 22 according to the image information, so that the leg power module 22 adjusts the position and angle of treading on the yeast 40 to obtain the yeast 40 with the target shape. In such a yeast-treading device 10, on the one hand, by monitoring the shape of the yeast 40 through the vision detection unit 3, the stability of the shape of the yeast 40 can be improved, the risk of the shape of the yeast 40 deviating from the target shape can be reduced, and the yield of yeast can be improved; on the other hand, by using the robot 2 to make yeast, the labor cost of yeast making can be effectively reduced and the yeast making efficiency of the yeast 40 can be improved.

[0080] In some embodiments, please refer to Figure 10 , Figure 10 This is a schematic diagram of the structure of the bending device 10 provided in some embodiments of this application. The frame 1 includes a fixed frame 11 and a movable frame 12. The movable frame 12 is movably disposed on the fixed frame 11 along the vertical Z direction, and the main body 21 is suspended on the movable frame 12.

[0081] The main body 21 is suspended on the movable frame 12 so that the robot 2 is suspended on the movable frame 12 as a whole. The movable frame 12 moves along the vertical Z direction, which drives the robot 2 to move along the vertical Z direction as a whole, so that the robot 2 moves closer to or further away from the yeast 40.

[0082] For example, such as Figure 10 As shown, the movable frame 12 includes a guide plate 123, and a guide groove is provided at the corner of the guide plate 123. The guide groove extends vertically along the Z direction. The fixed frame 11 has a column extending vertically along the Z direction. The column is inserted into the guide groove. The guide plate 123 moves vertically along the Z direction under the guidance of the column.

[0083] The fixed frame 11 can also be equipped with a limit block, which is located below the movable frame 12. The limit block is used to restrict the movable frame 12 from moving downward along the vertical Z direction, so as to keep the frame 1 from lifting the robot 2.

[0084] In this embodiment, the torso 21 is suspended on the movable frame 12, and the weight of the robot 2 is mainly borne by the frame 1. When stepping on the yeast 40, the robot 2's movement is determined by adjusting the stroke and orientation of its feet through the joint drive unit 221. Even if pressure sensors are installed at the joints to provide feedback on the stepping pressure, excessive foot stroke can easily lead to excessive stepping force due to feedback lag or monitoring failure, which is not conducive to the shaping of the yeast 40. By suspending the torso 21 on the movable frame 12 that is movable along the vertical Z direction, when the leg power module 22 steps on the yeast, the reaction force of the leg power module 22 can drive the movable frame 12 to move upward, thereby reducing the risk of the leg power module 22 stepping on the yeast and damaging the shape of the yeast 40, thus reducing the risk of yeast making failure and improving the yield rate of yeast.

[0085] In some embodiments, please refer to Figure 11 , Figure 11 This is a schematic diagram of the structure of the treading device 10 provided in some embodiments of this application. The frame 1 includes a buffer mechanism 13, which is connected to the fixed frame 11 and is configured to elastically support the movable frame 12.

[0086] The buffer mechanism 13 can be a spring, a buffer pad, etc. The buffer mechanism 13 elastically supports the movable frame 12 to reduce the impact force when the movable frame 12 moves downward and bends.

[0087] In this embodiment, by setting a buffer mechanism 13, the movable frame 12 can be elastically supported. During the downward movement of the movable frame 12, the buffer mechanism 13 can slow down the downward movement speed of the movable frame 12, thereby reducing the risk that the movable frame 12 will drive the leg power module 22 to impact the yeast 40 downward and damage the shape of the yeast 40, and improving the yield of yeast production.

[0088] In some embodiments, please continue to refer to Figure 11 And further refer to Figure 12 , Figure 12 for Figure 11A partial enlarged view of area A. The movable frame 12 includes a limiting plate 121, a connecting member 122, and a guide plate 123. The limiting plate 121 and the guide plate 123 are spaced apart vertically in the Z direction. The connecting member 122 connects the limiting plate 121 and the guide plate 123. The torso body 21 is connected to the guide plate 123. The guide plate 123 is guidedly connected to the fixed frame 11 to guide the guide plate 123 to move vertically in the Z direction. The buffer mechanism 13 includes a buffer plate 131 and an elastic buffer member 132. The buffer plate 131 includes a first end 1311 and a second end 1312 opposite to each other. The first end 1311 is hinged to the fixed frame 11, and the second end 1312 extends toward the bottom of the limiting plate 121. A sliding groove 1313 is provided on the surface of the buffer plate 131 opposite to the limiting plate 121. The sliding groove 1313 extends along the arrangement direction of the first end 1311 and the second end 1312. The elastic buffer 132 is located at the bottom of the buffer plate 131. The elastic buffer 132 includes a third end 1321 and a fourth end 1322 opposite to each other. The third end 1321 is hinged to the fixing frame 11 and spaced apart from the first end 1311. The fourth end 1322 is movably disposed in the slide groove 1313.

[0089] The limiting plate 121 cooperates with the buffer mechanism 13 to restrict the position of the robot 2, thereby keeping the robot 2 suspended on the frame 1. The limiting plate 121 can be located above the guide plate 123. The limiting plate 121 and the guide plate 123 are connected by a connector 122. Alternatively, both the limiting plate 121 and the guide plate 123 can be welded to the connector 122, or both can be bolted to the connector 122. The guide plate 123 is provided with a guide groove, and the vertical beam of the fixing frame 11 is engaged in the guide groove to guide the guide plate 123 to move vertically in the Z direction.

[0090] Both the first end 1311 of the buffer plate 131 and the third end 1321 of the elastic buffer member 132 are hinged to the fixing frame 11. The first end 1311 and the third end 1321 are spaced apart vertically in a Z-direction, with the third end 1321 located below the first end 1311. The second end 1312 of the buffer plate 131 extends obliquely downward toward the bottom of the limiting plate 121. When the limiting plate 121 moves downward, it first abuts against the upper surface of the buffer plate 131. The pressure on the buffer plate 131 is transmitted to the elastic buffer member 132, which elastically contracts. The fourth end 1322 of the elastic buffer member 132 slides within the groove 1313 to cushion the downward movement of the limiting plate 121. The elastic buffer member 132 can be a gas spring, a coil spring, or the like.

[0091] In this embodiment, by setting the buffer plate 131 to be hinged to the fixed frame 11, and the fourth end 1322 of the elastic buffer member 132 is movably set in the groove 1313 of the buffer plate 131, when the limiting plate 121 moves downward, the limiting plate 121 squeezes the buffer plate 131, and the buffer plate 131 is buffered by the elastic buffer member 132, thereby slowing down the downward movement speed of the limiting plate 121, thereby reducing the risk of the leg power module 22 moving downward too fast and damaging the structure of the yeast 40, and improving the yield of yeast production.

[0092] In some embodiments, please continue to refer to Figure 11 and Figure 12 And further refer to Figure 13 , Figure 13 This is an assembly diagram of a buffer plate 131 and an elastic buffer member 132 provided in some embodiments of this application. The buffer plate 131 includes a guide section 1314, an arc-shaped section 1315, and a support section 1316. The guide section 1314 is hinged to the fixing frame 11, and a groove 1313 is provided in the guide section 1314. The guide section 1314 is configured to compress the elastic buffer member 132 during downward movement under the limiting plate 121. The arc-shaped section 1315 is connected to the end of the guide section 1314 opposite to the fixing frame 11. The arc-shaped section 1315 forms a groove 13151. The limiting plate 121 is engaged in the groove 13151. The arc-shaped section 1315 is configured to compress the guide section 1314 and compress the elastic buffer member 132 during upward movement of the limiting plate 121, so as to allow the limiting plate 121 to disengage from the groove 13151. The support segment 1316 is connected to the end of the arc-shaped segment 1315 opposite to the guide segment 1314. The support segment 1316 is configured to support the limiting plate 121 when the limiting plate 121 is engaged in the groove 13151, thereby restricting the downward movement of the limiting plate 121. The end of the guide segment 1314 opposite to the arc-shaped segment 1315 is the first end 1311, and the end of the support segment 1316 opposite to the arc-shaped segment 1315 is the second end 1312.

[0093] The guide section 1314 of the buffer plate 131 can guide the movement of the limiting plate 121 so that when the limiting plate 121 moves downward, it is guided to a position where it is engaged with the arc-shaped section 1315. As the limiting plate 121 abuts against the guide section 1314 and moves downward, the guide section 1314 intensifies the compression of the elastic buffer member 132, thereby improving the buffering effect on the limiting plate 121.

[0094] The groove 13151 of the arc-shaped segment 1315 can hold the limiting plate 121. The arc-shaped segment 1315 and the support segment 1316 cooperate to stabilize the position of the limiting plate 121. When the leg power module 22 steps on the yeast 40, if the force of stepping on the yeast 40 is greater than the weight of the robot 2 and the movable frame 12, a compressive force will be generated between the movable frame 12 and the arc-shaped segment 1315. When the compressive force reaches a preset value, the arc-shaped segment 1315 drives the guide segment 1314 to compress the elastic buffer 132, so that the arc-shaped segment 1315 avoids the movable frame 12, thereby allowing the movable frame 12 to detach upward from the arc-shaped segment 1315. This structure is beneficial for adjusting the maximum pressure value of the robot 2 stepping on the yeast 40, balancing the improvement of yeast production quality and the reduction of the risk of damaging the yeast 40.

[0095] The support section 1316 extends downward toward the limiting plate 121. The support section 1316 can restrict the limiting plate 121 from moving further downward, thereby maintaining the hoisting of the robot 2 by the frame 1.

[0096] In this embodiment, by setting the guide section 1314 to be hinged to the fixed frame 11, and the slide groove 1313 is set on the guide section 1314, when the limiting plate 121 moves downward, the limiting plate 121 squeezes the guide section 1314, and the guide section 1314 compresses the elastic buffer 132, so that the guide section 1314 slows down the downward movement speed of the limiting plate 121 during the process of guiding the limiting plate 121 to the arc section 1315, thereby reducing the risk that the leg power module 22 will move downward too fast and damage the structure of the yeast 40. By setting the arc-shaped segment 1315, after the limiting plate 121 is engaged in the arc-shaped segment 1315, the elastic buffer 132 can resist the guide segment 1314, thereby restricting the limiting plate 121 within the arc-shaped segment 1315. When the pressure of the leg power module 22 is too high, the leg power module 22 pushes the limiting plate 121 upward, and the limiting plate 121 squeezes the groove wall of the groove 13151 of the arc-shaped segment 1315, thereby causing the guide segment 1314 to squeeze the elastic buffer 132, so that the leg power module 22 can move upward, reducing the risk of the yeast 40 being crushed and improving the yield of yeast production. By setting the support segment 1316, the support segment 1316 can support the limiting plate 121 when it is located in the groove 13151, reducing the risk of the limiting plate 121 detaching from the buffer mechanism 13 and falling off the frame 1, and improving the structural stability of the yeast-making device 10.

[0097] In some embodiments, please refer to Figure 14 , Figure 14This is an exploded view of the foot assembly 222 provided in some embodiments of this application. The leg power module 22 includes the foot assembly 222, and a plurality of pressure detection units 223 are provided on the bottom of the foot assembly 222. The plurality of pressure detection units 223 are used to acquire flexing force information of different areas of the bottom of the foot assembly 222. The control unit 4 is communicatively connected to the plurality of pressure detection units 223, and the control unit 4 is used to control the action of a plurality of joint drive units 221 according to the flexing force information, so that the leg power module 22 treads the yeast 40.

[0098] The number of pressure detection units 223 can be two, three, four, five, or more. The pressure detection units 223 are distributed in different areas on the bottom of the foot assembly 222. The control unit 4 can communicate with multiple pressure detection units 223 via a Bluetooth module; alternatively, the control unit 4 can communicate with multiple pressure detection units 223 via wires. The pressure detection units 223 can be strain gauge pressure sensors, piezoresistive pressure sensors, capacitive pressure sensors, etc. The pressure detection units 223 can be in direct or indirect contact with the yeast 40. For example, the foot assembly 222 includes a foot 2221 and a separator membrane 2222. Multiple receiving grooves are provided on the bottom of the foot 2221, and the pressure detection units 223 are disposed within these grooves. The pressure detection units 223 can be connected to the control unit 4 via Bluetooth. The separator membrane 2222 is disposed on the bottom of the foot 2221 and seals the multiple receiving grooves. The separator membrane 2222 is made of a soft material, and the pressure detection units 223 sense the force applied when stepping on the yeast 40 through the separator membrane 2222.

[0099] In this embodiment, when making the yeast starter 40, in addition to considering its shape, the density and moisture content of the yeast starter 40 are also important factors. By setting up a pressure detection unit 223 to sense the treading force information, the density of the yeast starter 40 at different locations can be determined, which is beneficial for analyzing the moisture content and compactness of the yeast starter 40, thus improving the quality of yeast starter production. By setting multiple pressure detection units 223 in the foot assembly 222, the obtained treading force information helps to determine the flatness of the yeast starter 40 and the density of the yeast starter 40 in a fixed area. Multiple joint drive units 221 can adjust the treading position and treading angle of the foot assembly 222 according to the treading force of different areas of the foot assembly 222 obtained by the multiple pressure detection units 223, thereby treading on the target position of the yeast starter 40, which is beneficial for obtaining yeast starter 40 with the target shape and density, thus improving the quality of yeast starter production.

[0100] In some embodiments, please continue to refer to Figure 14 Multiple pressure detection units 223 are arrayed at the bottom of the foot assembly 222.

[0101] Multiple pressure detection units 223 are arranged in multiple rows and columns on the bottom of the foot assembly 222. The number of pressure detection units 223 in each row may be equal or unequal. The number of pressure detection units 223 in each column may be equal or unequal.

[0102] In this embodiment, by arraying multiple pressure detection units 223, the pressure detection units 223 are positioned more evenly at the bottom of the foot component 222, resulting in more uniform pressure feedback at the bottom of the foot component 222.

[0103] In some embodiments, please refer to Figure 15 , Figure 15 This is an assembly diagram of the lower leg assembly 224 and foot assembly 222 provided in some embodiments of this application. The leg power module 22 includes the lower leg assembly 224 and the foot assembly 222, which are movably connected to the lower leg assembly 224. Multiple joint drive units 221 include a first drive unit 2211, which is drively connected to the foot assembly 222. The first drive unit 2211 is used to drive the foot assembly 222 to swing up and down and left and right relative to the lower leg assembly 224.

[0104] The first drive unit 2211 can be located at the ankle joint where the lower leg assembly 224 and the foot assembly 222 connect, or it can be located at the foot assembly 222 or the lower leg assembly 224. The power output end of the first drive unit 2211 can be directly connected to the foot assembly 222; or the power output end of the first drive unit 2211 can be indirectly connected to the foot assembly 222.

[0105] In this embodiment, by setting the first driving unit 2211 to drive the foot assembly 222 to swing up and down and left and right relative to the lower leg assembly 224, the first driving unit 2211 can adjust the stepping angle of the foot assembly 222 according to the image information provided by the vision detection unit 3, reducing the difficulty of bending. In addition, the flexible adjustment of the bottom of the foot assembly 222 is beneficial to set the bottom of the foot assembly 222 to a larger area, so as to accommodate the stepping area and stepping accuracy, and improve the bending efficiency.

[0106] In some embodiments, please continue to refer to Figure 15 The first drive unit 2211 includes two drive members 22111 disposed on the lower leg assembly 224. One drive member 22111 is connected to the foot assembly 222 via a first connecting rod 22112, and the other drive member 22111 is connected to the foot assembly 222 via a second connecting rod 22113. The first connecting rod 22112 and the second connecting rod 22113 are respectively located on both sides of the foot assembly 222 in the left-right direction.

[0107] The rotation of the two drive components 22111 can drive the first link 22112 and the second link 22113 to move synchronously upward or downward, thereby causing the foot assembly 222 to tilt up or drop. The rotation of one of the two drive components 22111 can drive one of the first link 22112 and the second link 22113 to tilt up or drop, thereby causing the foot assembly 222 to swing left and right.

[0108] In this embodiment, by setting two drive members 22111 on the lower leg assembly 224, the two drive members 22111 can be set away from the foot assembly 222, thereby reducing the risk of corrosion damage to the drive members 22111 from contact with the yeast 40 and improving the service life of the leg power module 22.

[0109] In some embodiments, please refer to Figure 16 and Figure 17 , Figure 16 This is a schematic diagram of the structure of the leg power module 22 provided in some embodiments of this application; Figure 17 This is an assembly diagram of the joint drive unit 221 provided in some embodiments of this application. The leg power module 22 includes a thigh assembly 225 and a lower leg assembly 224 connecting the thigh assembly 225 and the foot assembly 222; the multiple joint drive units 221 include a second drive unit 2212 and a third drive unit 2213, the second drive unit 2212 is used to drive the lower leg assembly 224 to swing back and forth relative to the thigh assembly 225, and the third drive unit 2213 is used to drive the thigh assembly 225 to swing left and right relative to the torso body 21.

[0110] The second drive unit 2212 can be located at the knee joint between the lower leg assembly 224 and the thigh assembly 225, or it can be located at either the lower leg assembly 224 or the thigh assembly 225. The third drive unit 2213 can be located at the hip joint between the thigh assembly 225 and the torso body 21, or it can be located at either the thigh assembly 225 or the torso body 21.

[0111] In this embodiment, by setting the second drive unit 2212 to drive the lower leg assembly 224 to swing back and forth relative to the thigh assembly 225, and the third drive unit 2213 to drive the thigh assembly 225 to swing left and right relative to the torso body 21, the foot assembly 222 can move back and forth and left and right, which makes it easier to adjust the bending position of the leg power module 22.

[0112] In some embodiments, please continue to refer to Figure 16 and Figure 17The multiple joint drive units 221 also include a fourth drive unit 2214 and a fifth drive unit 2215. The fourth drive unit 2214 is used to drive the thigh assembly 225 to rotate circumferentially relative to the torso body 21, and the fifth drive unit 2215 is used to drive the thigh assembly 225 to swing back and forth relative to the torso body.

[0113] The fourth drive unit 2214 can be located at the hip joint connection between the thigh assembly 225 and the torso body 21, or it can be located on the thigh assembly 225 or on the torso body 21. The fifth drive unit 2215 can be located at the hip joint connection between the thigh assembly 225 and the torso body 21, or it can be located on the thigh assembly 225 or on the torso body 21.

[0114] In this embodiment, by setting the fourth drive unit 2214 to drive the thigh assembly 225 to rotate circumferentially relative to the torso body 21, and the fifth drive unit 2215 to drive the thigh assembly 225 to swing back and forth relative to the torso body, it is beneficial to further adjust the bending angle of the foot assembly 222 and expand the bending area of ​​the foot assembly 222, thereby improving the performance of the bending device 10.

[0115] In some embodiments, please refer to Figure 18 , Figure 18 This is an assembly diagram of the joint drive unit 221 provided in some embodiments of this application. The plurality of joint drive units 221 also include a fourth drive unit 2214 and a fifth drive unit 2215. The fourth drive unit 2214 is used to drive the thigh assembly 225 to swing back and forth relative to the torso body. The thigh assembly 225 includes a first thigh portion 2251 and a second thigh portion 2252. The second thigh portion 2252 connects the first thigh portion 2251 and the lower leg assembly 224. The fifth drive unit 2215 is used to drive the second thigh portion 2252 to rotate circumferentially relative to the first thigh portion 2251.

[0116] The fourth drive unit 2214 can be disposed at the hip joint connection between the thigh assembly 225 and the torso body 21, or it can be disposed at either the thigh assembly 225 or the torso body 21. The fifth drive unit 2215 can be disposed at the connection between the first thigh 2251 and the second thigh 2252, or it can be disposed at either the first thigh 2251 or the second thigh 2252.

[0117] In this embodiment, by setting four drive units to drive the thigh assembly 225 to swing back and forth relative to the torso, and the fifth drive unit 2215 to drive the second thigh 2252 to rotate circumferentially relative to the first thigh 2251, it is convenient to realize the circumferential rotation of the foot assembly 222's back and forth swing, which is beneficial to further adjust the bending angle of the foot assembly 222 and expand the bending area of ​​the foot assembly 222, thereby improving the performance of the bending device 10.

[0118] This application provides a koji-making production line 100, including a conveyor belt 20 and a koji-making device 10 provided in any of the above embodiments. The conveyor belt 20 is used to transport a koji mold 30 containing koji 40. A robot 2 is located above the conveyor belt 20 and is used to tread on the koji 40 located in the koji mold 30.

[0119] Robot 2 is located above conveyor belt 20. The yeast mold 30 is transported to the area below robot 2 via conveyor belt 20, and robot 2, which is suspended on frame 1, steps on the yeast 40 to make yeast.

[0120] In this embodiment, by setting up a conveyor belt 20 to transport the yeast mold 30, the robot 2 is located above the conveyor belt 20. The conveyor belt 20 transports the yeast mold 30 to the bottom of the conveyor belt 20 for treading, which helps to improve the efficiency of yeast making and reduce the difficulty of yeast making.

[0121] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0122] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0123] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0124] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0125] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A control method for a treadmill robot, characterized in that, The method, applied to the controller of the treading robot, includes: The surface of the curved mold is divided into multiple grids; For each grid cell, the target parameter matching degree is calculated. If the target parameter matching degree is less than the matching degree threshold, the corresponding grid cell is added to the trampling sequence. Perform one round of trampling on all grids in the trampling sequence; Repeat the above process until the target parameter matching degree of all grids is greater than or equal to the matching degree threshold, or the number of pedaling cycles exceeds the set threshold, then end the pedaling process.

2. The method as described in claim 1, characterized in that, The target parameter matching degree includes the high matching degree; The method for calculating the high degree of matching includes: Use a depth camera to obtain a height map, and get the height h(x, y) of the grid. The high degree of matching is: ； Among them, h * (x, y) represents the optimal block height, where x and y are the grid plane coordinates.

3. The method as described in claim 2, characterized in that, The target parameter matching degree includes the density matching degree; The method for calculating the density matching degree includes: The force spectrum is recorded during the treading process; where the force spectrum is the change of the applied force over time during the treading process. The force spectrum matching degree is: ； Where F(x, y, t) is the force spectrum; F * (x, y, t) represents the empirically optimal force spectrum, t represents the current time, and T represents the time taken for a single pedal stroke; The force spectrum matching degree is used as the density matching degree.

4. The method as described in claim 3, characterized in that, The step of stepping on all grids in the stepping sequence in one round includes: Based on the position cost function, velocity cost function, acceleration cost function, and robot dynamics model, force-position hybrid control of pedaling is achieved.

5. The method as described in claim 4, characterized in that, The location cost function is: ； The speed cost function is: ； Where q is the joint space coordinate, q d Let x be the target space coordinates of the joint. i Let x be the task space coordinate. i d J represents the spatial coordinates of the mission target. i For the task Jacobian matrix, W i As the task weight, W q Here, N represents the joint weight, and N represents the total number of tasks.

6. The method as described in claim 5, characterized in that, The acceleration cost function is: ； The robot's dynamics model is as follows: ； ； ； Where, f c For the foot end to bear the force, f d For the target at the foot, J c Let τ be the Jacobian matrix of the forces acting on the foot. d For the target torque of the joint, W f H is the force weight at the foot, C is the inertia matrix, G is the centrifugal force matrix, and W is the gravity term. τ For joint torque weighting, The range of force applied to the foot. The target torque of the joint is the minimum value. This represents the maximum target torque of the joint.

7. The method as described in claim 6, characterized in that, The stepping control, which achieves force-position hybrid control based on position cost function, velocity cost function, acceleration cost function, and robot dynamics model, includes: Non-contact phase: For the stepped grid (x, y), control the foot to move to coordinate (x, y, h), and simultaneously apply force f to the foot. c The force in the x, y, and z directions is limited to 0 until the foot touches the curved block; where h is the height of the foot. In the non-contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

8. The method as described in claim 7, characterized in that, The stepping control, which achieves force-position hybrid control based on position cost function, velocity cost function, acceleration cost function, and robot dynamics model, further includes: Contact phase: After the foot contacts the curved block, the force f applied to the foot will be... c The force in the x and y directions is limited to 0, controlling the force f on the foot. c The force in the z-direction is: ； Among them, K F K h This is the proportionality coefficient; During the contact phase, the optimization problem is solved using the position cost function, velocity cost function, acceleration cost function, and robot dynamics model to obtain the joint target spatial coordinates q. d Differentiation of joint target space coordinates and the joint target torque τ d And convert it into motor position. Motor speed and current command .

9. The method as described in claim 8, characterized in that, The foot of the treading robot is equipped with multiple pressure detection units at different positions on the bottom of its feet.

10. The method as described in claim 9, characterized in that, The stepping control, which achieves force-position hybrid control based on position cost function, velocity cost function, acceleration cost function, and robot dynamics model, further includes: Determine whether the difference between the average pressures on both sides is greater than the first lateral deviation threshold and the duration is greater than the corresponding threshold. If so, continue to determine whether the difference between the average pressures on both sides is less than the second lateral deviation threshold. If the difference in average pressure between the two sides is less than the second lateral deviation threshold, then control the foot's fine-tuning angle: ； Where, k θ PL and PR are proportional parameters, and PL and PR are the average values ​​of the pressure on the opposite side. If the average pressure on both sides is greater than or equal to the second lateral deviation threshold, then reduce the force f on the foot. c Apply force in the z-direction and control the foot to return to a safe height, then replan the pedaling motion.

11. An electronic device, characterized in that, include: A processor and a memory, the memory storing machine-readable instructions executable by the processor, which, when executed by the processor, perform the method as described in any one of claims 1-10.

12. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-10.

Images