Posture generation system, posture generation method, and program

Abstract

To provide a posture generation system, posture generation method, and program for suitably generating the posture of a human model during the transport of goods. [Solution] A posture generation system according to one aspect of the present disclosure comprises a transport item identification unit, a reference position determination unit, and a posture generation unit. The transport item identification unit identifies the transport item that the human model will be transporting in the simulation. The reference position determination unit determines at least one first reference position on the human model based on the transport item, and further determines at least one second reference position corresponding to the feet of the human model based on walking information related to the human model's walking. The posture generation unit generates the posture of the human model based on the determined first and second reference positions.

Description

Technical Field

[0001] The present invention relates to a posture generation system, a posture generation method, and a program.

Background Art

[0002] In work efficiency improvement and risk management, it is important to evaluate the fatigue level or posture stability during work. There is a method that uses simulation of the work environment for evaluating fatigue level or posture stability, etc. Here, high-precision simulation for evaluating the work environment requires reproduction of an accurate work posture in a human model. Patent Document 1 describes a posture estimation device that can estimate an initial posture quickly and with high precision.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The posture estimation device described in Patent Document 1 performs detailed fitting and optimization processing based on an image including the calculated initial posture and a detailed human model to finally estimate the initial posture. Thereby, the initial posture can be estimated with higher precision. However, since the posture estimation device described in Patent Document 1 detects the contact between the object components of an object fixed to the environment such as a chair or a desk and the human body parts, it is not suitable for estimating the posture in an environment where there is little contact with the articles fixed to the environment.

[0005] Incidentally, if the posture of a human model during the handling of goods can be appropriately generated, it can be used in combination with machine learning to design an efficient work environment. However, generally, contact with objects fixed to the environment is infrequent during the handling of goods. Therefore, the posture estimation device described in Patent Document 1 is difficult to apply to generating the posture of a human model during movement, such as when handling goods.

[0006] In view of the above-mentioned problems, this disclosure provides a posture generation system, etc., that suitably generates the posture of a human model during the transport of goods. [Means for solving the problem]

[0007] A posture generation system according to one aspect of the present disclosure comprises a transport item identification unit, a reference position determination unit, and a posture generation unit. The transport item identification unit identifies the transport item carried by the human model in the simulation. The reference position determination unit determines at least one first reference position on the human model based on the transport item, and further determines at least one second reference position corresponding to the feet of the human model based on walking information related to the human model's walking. The posture generation unit generates the posture of the human model based on the determined first and second reference positions.

[0008] In the posture generation system described above, the first reference position may be a reference position corresponding to the fingertips of the human model. In this case, the reference position determination unit may determine a second reference position based on the movement speed of the human model, and the posture generation unit may determine the positions of the waist and head of the human model based on the first and second reference positions.

[0009] In the posture generation system described above, the posture generation unit may determine the position of the head based on the height of the first reference position.

[0010] A posture generation method according to one aspect of this disclosure involves a computer performing the following processes: The computer identifies the object being carried by the human model. Based on the object, the computer determines at least one first reference position on the human model, and further determines at least one second reference position corresponding to the feet of the human model based on walking information relating to the human model's gait. Based on the determined first and second reference positions, the computer generates the posture of the human model.

[0011] A program according to one aspect of this disclosure causes a computer to perform the following processes: The computer identifies the goods carried by the human model. Based on the goods, the computer determines at least one first reference position on the human model, and based on gait information relating to the human model's walking, further determines at least one second reference position corresponding to the human model's feet. Based on the determined first and second reference positions, the computer generates a posture for the human model. [Effects of the Invention]

[0012] This disclosure provides a posture generation system, posture generation method, and program for suitably generating the posture of a human model during the transport of goods. [Brief explanation of the drawing]

[0013] [Figure 1] This is a block diagram of the posture generation system according to Embodiment 1. [Figure 2] This is a flowchart of the posture generation method according to Embodiment 1. [Figure 3] This is a flowchart of the posture generation method according to Embodiment 2. [Figure 4] This is a flowchart of the head position calculation method according to Embodiment 2. [Figure 5] This is a schematic diagram of the head position calculation method according to Embodiment 2. [Figure 6] This is a block diagram illustrating the hardware configuration of a computer. [Modes for carrying out the invention]

[0014] The present invention will be described below through embodiments of the invention, but the invention claimed is not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential as means of solving the problem. For clarity of explanation, the following descriptions and drawings have been omitted and simplified as appropriate. In each drawing, the same elements are denoted by the same reference numerals, and redundant explanations have been omitted where necessary.

[0015] <Embodiment 1> Referring to Figure 1, the posture generation system 10 according to Embodiment 1 will be described. Figure 1 is a block diagram of the posture generation system 10 according to Embodiment 1. The posture generation system 10 comprises a transported item identification unit 101, a reference position determination unit 102, and a posture generation unit 103. The posture generation system 10 generates the posture of a human model that transports the transported item in the simulation. Here, the posture generation system 10 calculates the posture of the human model from a first reference position and a second reference position. The first reference position is a reference position related to the transported item. The second reference position is a reference position corresponding to the feet of the human model. Note that the feet are not limited to the toes, but may refer to the knee, above the knee, or below the knee, etc.

[0016] The item identification unit 101 identifies the item being carried by the human model. The item is, for example, materials, a box, or a tool. The item identification unit 101 may also identify the type of shape of the item. The shape of the item is, for example, a box, a rod, a plate, or the shape of a tool with a handle. The item may also have feature information such as the presence or absence of a handle, weight, or center of gravity. Here, the item identification unit 101 may use recognition software. The recognition software may be programmed to perform a recognition method, or it may have learned a recognition method through machine learning. Alternatively, the item identification unit 101 may use artificial intelligence (AI) for recognition that has been trained in advance.

[0017] Note that the transported item specifying unit 101 may store, as a database, the feature information of a plurality of transported items input in advance. Further, the transported item specifying unit 101 may extract feature information from the shape of the transported item, for example.

[0018] The reference position determination unit 102 determines at least one first reference position in the human model based on the transported item. For example, the reference position determination unit 102 determines the contact position between the human model and the transported item as the first reference position.

[0019] The reference position determination unit 102 further determines at least one second reference position corresponding to the feet of the human model based on the walking information related to the walking of the human model. The walking information includes, for example, the moving speed of the human model, the stride, the slope of the ground, or the presence or absence of a step.

[0020] The posture generation unit 103 generates the posture of the human model based on the determined first reference position and second reference position. For example, the posture generation unit 103 generates the posture of the human model using inverse kinematics calculation. Inverse kinematics calculation is a method that assumes that a predetermined reference position on the human model is fixed, and calculates the positions of the remaining undetermined parts by inverse calculation so as not to be dynamically unnatural on that premise.

[0021] According to this, the posture generation system 10 can determine the reference position on the human model based on the transported item and generate the posture of the human model. Therefore, the posture generation system 10 can suitably generate the posture of the human model even when carrying an item with little contact with an object fixed to the environment.

[0022] FIG. 2 is a flowchart of the posture generation method according to Embodiment 1. The posture generation method according to Embodiment 1 includes steps S11 to S14.

[0023] In step S11, the posture generation system 10 identifies the cargo that the human model is carrying. The posture generation system 10 only needs to be able to identify the type of shape of the cargo. In step S12, the posture generation system 10 determines at least one first reference position of the human model based on the cargo. If characteristic information of the identified cargo can be obtained, the posture generation system 10 may also determine the relationship between the cargo and the reference position based on the characteristic information of the cargo.

[0024] In step S13, the posture generation system 10 further determines at least one second reference position corresponding to the feet of the human model based on gait information relating to the human model's walking. In step S14, the posture generation system 10 generates the posture of the human model based on the first reference position and the second reference position. Here, the posture generation system 10 generates the posture of the human model using, for example, inverse kinematics calculations.

[0025] As explained above, the posture generation system 10 generates the posture of a human model based on the object being carried by the human model in the simulation. This allows for the appropriate calculation of the posture of the human model from a first reference position related to the object and a second reference position corresponding to the human model's feet. Therefore, by using the posture generation system 10, posture can be appropriately simulated when transporting objects in which contact with objects fixed to the environment is infrequent.

[0026] The attitude generation system 10 may also have a processor and a memory device, although these are not shown in the diagram. The memory device of the attitude generation system 10 may include, for example, a memory device that includes non-volatile memory such as flash memory or an SSD (Solid State Drive). In this case, the memory device stores a computer program (hereinafter also simply referred to as a program) for executing the above-described method. The processor loads the computer program from the memory device into a buffer memory such as DRAM (Dynamic Random Access Memory) and executes the program.

[0027] Each component of the attitude generation system 10 may be implemented with dedicated hardware. Furthermore, some or all of each component may be implemented by general-purpose or dedicated circuits, processors, etc., or combinations thereof. These may be implemented by a single chip or by multiple chips connected via a bus. Some or all of each component of each device may be implemented by a combination of the aforementioned circuits, etc., and programs. Processors include CPUs (Central Processing Units), GPUs (Graphics Processing Units), FPGAs (Field-Programmable Gate Arrays), etc. Also, at least a portion of the processing performed by the attitude generation system 10 may be provided as SaaS (Software as a Service). The descriptions of the configurations described herein may also apply to other devices or systems described below in this disclosure.

[0028] <Embodiment 2> Figure 3 is a flowchart of the posture generation method according to Embodiment 2. The posture generation system according to Embodiment 2 has the same configuration as the posture generation system 10 in Figure 1. Therefore, a description of the configuration of the posture generation system that performs the same processing as the posture generation system 10 is omitted. In the posture generation method according to Embodiment 2, the reference position determination unit determines the second reference position P2 based on the movement speed of the human model. The posture generation unit then determines the positions of the waist and head of the human model 21 based on the first reference position P1 and the second reference position P2 and generates the posture. The posture generation method according to Embodiment 2 includes steps S21 to S26.

[0029] In step S21, the transport item identification unit of the posture generation system identifies the transport item 22 to be transported by the human model 21. Here, the human model 21 includes the positions of the hands, feet, waist, and head. The transport item identification unit also identifies the type of shape and characteristic information of the transport item 22. Characteristic information includes information such as the presence or absence of a handle, weight, or center of gravity. The transport item identification unit may store the type of shape and characteristic information in advance and read the characteristic information by identifying the type of shape of the transport item 22.

[0030] In step S22, the reference position determination unit of the posture generation system determines a first reference position P1 based on the transported object 22. Here, the first reference position P1 is the position corresponding to the fingertips of the human model 21 that holds the transported object 22. There may be two first reference positions P1, one for each hand. The relationship between the transported object 22 and the first reference position P1 is determined based on the shape and characteristic information of the transported object 22. In other words, the reference position determination unit changes the way the human model 21 holds the transported object 22. Note that the fingertips refer to the part of the hand from the wrist downwards and may correspond to the wrist, palm, or fingertips of the human model 21.

[0031] For example, if the item being transported 22 is a tool with a handle, the position of the tool's handle is set as the first reference position P1, representing the position of the fingertips of the human model 21. If the item being transported 22 is a box without a handle, the positions of the two opposing corners of the bottom surface are set as the first reference positions P1, representing the positions of the fingertips of the human model's hands. The reference position determination unit may pre-define at least one first reference position P1 on the human model for each type of shape of the item being transported 22, store it in a database, and retrieve it as needed. Alternatively, the reference position determination unit may determine at least one first reference position P1 on the human model each time, reflecting feature information.

[0032] In step S23, the reference position determination unit of the posture generation system determines a second reference position P2 based on the origin position P0 and movement speed of the human model 21. Here, the second reference position P2 is the position corresponding to the toes of the human model 21. Two second reference positions P2 may be set, one for each foot. The toes refer to the part of the foot from the ankle bone downwards and may correspond to the fingertips, heels, or ankles of the human model 21. The origin position P0 is positional information indicating the position of the human model 21 in the simulation. The origin position P0 is, for example, the center position of the toes of both feet of the human model. The origin position P0 may also be the center of gravity of the human model 21 on the ground plane in the simulation.

[0033] The reference position determination unit is set, for example, so that the distance from the origin position P0 to the second reference position P2 increases as the walking speed increases. This allows the posture generation system to reflect the widening of stride length according to walking speed, and to generate a posture with foot positions that are closer to reality.

[0034] Furthermore, the reference position determination unit may adjust the second reference position P2 based on the weight of the transported item. For example, the reference position determination unit adjusts the distance from the origin position P0 to the second reference position P2 so that the greater the weight of the transported item, the shorter the distance. This allows the posture generation system to reflect the change in stride length according to the weight of the transported item.

[0035] In step S24, the posture generation unit of the posture generation system calculates the waist position P3 of the human model 21 from the first reference position P1 and the second reference position P2. Alternatively, the posture generation unit may calculate the waist position P3 from either the first reference position P1 or the second reference position P2. Here, the posture generation unit uses inverse kinematics calculations. The posture generation unit may also use a value predetermined with respect to the origin position P0 as the waist position P3.

[0036] In step S25, the posture generation unit of the posture generation system calculates the head position P4 of the human model 21 from the first reference position P1, the second reference position P2, and the waist position P3. Alternatively, the posture generation unit may calculate the head position P4 of the human model 21 from the first reference position P1 and the waist position P3. Figure 4 is a flowchart of the head position calculation method according to Embodiment 2. The head position calculation method includes substeps S251 to S253.

[0037] In substep S251, the posture generation unit obtains the height X of the first reference position P1, which indicates the hand position. In substep S252, the posture generation unit calculates the anterior-posterior position Y of the head relative to the first reference position P1. Y is calculated by Y = aX + b, where X is the height of the first reference position P1, and a and b are predetermined coefficients.

[0038] In substep S253, the posture generation unit calculates the height Z of the head position P4 such that the distance L between the waist position P3 and the head position P4 remains constant. Based on this, the posture generation unit determines the head position P4 based on the height X of the first reference position P1. As described above, the posture generation unit can calculate the head position P4 through substeps S251 to S253.

[0039] Figure 5 is a schematic diagram of the head position calculation method according to Embodiment 2. Note that in Figure 5, the right hand and right foot of the human model 21 are omitted for simplicity. In Figure 5, the human model 21 is holding the transported item 22. Here, the relative front-to-back position Y of the head position P4 as seen from the first reference position P1 is determined in substep S252 in proportion to the height X of the first reference position P1. That is, the lower the height X of the first reference position P1, the closer the head position P4 is to the first reference position P1 in the front-to-back direction. Also, the higher the height X of the first reference position P1, the further the head position P4 is from the first reference position P1 in the front-to-back direction. At this time, the distance L between the waist position P3 and the head position P4 does not change.

[0040] According to this, the posture generation system can represent the movement of bending forward when lifting an object and leaning back when holding the object high. Therefore, the posture generation system can more effectively calculate the posture of a human model.

[0041] Referring again to Figure 3, in step S26, the posture generation unit calculates the positions of other body parts based on inverse kinematics. These other body parts include, but are not limited to, the shoulders, elbows, and knees. In this way, the posture generation system can generate the posture of the human model 21 based on the first reference position P1 and the second reference position P2. The posture generation unit may further calculate the positions of equipment worn by the human model. It is preferable that the first reference position P1 and the second reference position P2 be positions at the extremities of the body. In this way, the posture generation system can perform inverse calculations with high accuracy.

[0042] As explained above, the posture generation system generates the posture of a human model based on the object being carried and the human model's movement speed. This allows the posture generation system to more effectively simulate posture during object transport. Therefore, users of the posture generation system can accurately simulate transport work and effectively evaluate fatigue levels, posture stability, and other factors during transport work.

[0043] <Example hardware configuration> The following describes examples of how each functional configuration of the attitude generation system in this disclosure can be realized through a combination of hardware and software.

[0044] Figure 6 is a block diagram illustrating the hardware configuration of a computer. The attitude generation system in this disclosure can realize the above-described functions using a computer 500 including the hardware configuration shown in the figure. The computer 500 may be a portable computer such as a smartphone or tablet terminal, or a stationary computer such as a PC. The computer 500 may be a dedicated computer designed to realize each device, or it may be a general-purpose computer. The computer 500 can realize the desired functions by installing a predetermined application.

[0045] Computer 500 includes a bus 502, a processor 504, memory 506, a storage device 508, an input / output interface (I / F) 510, and a network interface (I / F) 512. Bus 502 is a data transmission path for the processor 504, memory 506, storage device 508, input / output interface 510, and network interface 512 to send and receive data to and from each other. However, the method of connecting the processor 504 and other components to each other is not limited to bus connection.

[0046] Processor 504 is a processor such as a CPU, GPU, or FPGA. Memory 506 is main memory implemented using RAM (Random Access Memory), etc.

[0047] The storage device 508 is an auxiliary storage device implemented using a hard disk, SSD, memory card, or ROM (Read Only Memory). The storage device 508 stores a program for realizing a desired function. The processor 504 reads this program into memory 506 and executes it to realize each functional component of each device.

[0048] The input / output interface 510 is an interface for connecting the computer 500 to input / output devices. For example, input devices such as keyboards and output devices such as display devices are connected to the input / output interface 510. The network interface 512 is an interface for connecting the computer 500 to a network.

[0049] It should be noted that the present invention is not limited to the embodiments described above, and can be modified as appropriate without departing from the spirit of the invention. For example, the posture generation system may automatically determine the position of the handtips according to the weight set for the transported object. Furthermore, the posture generation system may also be compatible with transported objects other than those held by the hands. Specifically, if the transported object is something that is carried on the shoulders, the posture generation system may include the position of the shoulders or back in the human model and generate the posture using the position of the shoulders or back that are in contact with the transported object as the reference position. [Explanation of Symbols]

[0050] 10. Posture Generation System 101 Item Identification Section 102 Reference position determination section 103 Posture generation section 21 models 22 Transported goods 500 Computers Bus 502 504 Processors 506 memory 508 Storage Devices 510 Input / Output Interfaces 512 Network Interfaces P0 Origin position P1 1st reference position P2 2nd reference position P3 Waist position P4 head position X Height Y front and back position Z Height L distance

Claims

1. It comprises a transport item identification unit, a reference position determination unit, and a posture generation unit, The aforementioned transport item identification unit identifies the transport items that the human model will be transporting in the simulation. The aforementioned reference position determination unit is Based on the transported goods, at least one first reference position in the human model is determined. Based on the gait information relating to the walking of the human model, at least one second reference position corresponding to the foot of the human model is further determined. The posture generation unit generates the posture of the human model based on the determined first reference position and second reference position. Posture generation system.

2. The first reference position is a reference position corresponding to the fingertips of the human model, The reference position determination unit determines the second reference position based on the movement speed of the human model. The posture generation unit determines the positions of the waist and head in the human model based on the first reference position and the second reference position. The posture generation system according to claim 1.

3. The posture generation unit determines the position of the head based on the height of the first reference position. The posture generation system according to claim 2.

4. Computers Identify the items that the human model will be carrying in the simulation. Based on the transported goods, at least one first reference position in the human model is determined. Based on the gait information relating to the walking of the human model, at least one second reference position corresponding to the foot of the human model is further determined. Based on the determined first reference position and second reference position, the posture of the human model is generated. Posture generation method.

5. On the computer, Identify the items that the human model will be carrying in the simulation. Based on the transported goods, at least one first reference position in the human model is determined. Based on the gait information relating to the walking of the human model, at least one second reference position corresponding to the foot of the human model is further determined. Based on the determined first reference position and second reference position, the posture of the human model is generated. A program that makes something happen.

Images