JP2022154770A - Control device for industrial vehicle, industrial vehicle and control program for industrial vehicle - Google Patents

Abstract

To provide a control device for an industrial vehicle that can accurately determine the transition of a driving state of the industrial vehicle, the industrial vehicle and a control program for the industrial vehicle.SOLUTION: A driving state determination unit 31 comprises: a transition condition determination unit 31A that determines a first driving state by comparing a detection result of a detection unit and a transition condition; and a statistical processing determination unit 31B that performs statistical processing on the detection result of the detection unit 32 and determines a second driving state based on modeled data. The driving state determination unit 31 settles the driving state determination result by comparing the determination result by the transition condition determination unit 31A and the determination result by the statistical processing determination unit 31B. In this way, the driving state determination unit 31 can determine the driving state with greater accuracy by comparing not only the determination result of the first driving state using the transition condition but also the determination result of the second driving state using the statistical processing.SELECTED DRAWING: Figure 2

Description

The present invention relates to an industrial vehicle control device, an industrial vehicle, and an industrial vehicle control program.

2. Description of the Related Art As a conventional control device for industrial vehicles, for example, the technology described in Patent Document 1 is known. The control device described in Patent Literature 1 is mounted on an industrial vehicle. The control device includes a driving state determination unit that determines the driving state of the industrial vehicle based on a state transition model in which the driving states of the industrial vehicle are classified. The operating state determination unit determines the operating state based on the progress state and the operating state of the cargo handling device.

JP 2019-189435 A

Here, in the field of industrial vehicles, by determining the driving state of the industrial vehicle of the control device, the determination result can be effectively used in various forms such as work support, grasping individual differences of the driver, autonomous driving, etc. can be used. Therefore, it has been required to further improve the accuracy of determination when determining the driving state of the industrial vehicle by the control device.

An object of the present invention is to provide a control device for an industrial vehicle, an industrial vehicle, and a control program for the industrial vehicle, which can accurately determine the transition of the operating state of the industrial vehicle.

A control device for an industrial vehicle according to an aspect of the present invention is a control device for an industrial vehicle including a vehicle and a cargo handling device, comprising: a detection unit that detects operation information based on operations of various operation means in the industrial vehicle; a driving state determination unit that determines the driving state of the industrial vehicle based on the state transition model in which the driving state of the industrial vehicle is classified, and the driving state determination unit compares the detection result of the detection unit with the transition condition. Thus, a transition condition determination unit that determines the first operating state, and a statistical processing determination unit that performs statistical processing on the detection result of the detection unit and determines the second operating state based on the modeled data. , and determines the driving state determination result by comparing the determination result by the transition condition determination unit and the determination result by the statistical processing determination unit.

In the industrial vehicle control device according to the aspect of the present invention, the driving state determination unit determines the driving state of the industrial vehicle based on a state transition model in which the driving states of the industrial vehicle are classified. In this manner, the driving state determination unit makes a determination based on the state transition model, making it easier to determine the driving state of the industrial vehicle. Here, the present inventors found the following knowledge as a result of earnest research. That is, when operation information based on operation of various operation means in industrial vehicles is detected and statistical processing is performed on the detection results, the processing results can be used to improve the accuracy of determining the driving state. I found something. Therefore, the driving state determination unit compares the detection result of the detection unit with the transition condition to perform statistical processing on the transition condition determination unit that determines the first driving state and the detection result of the detection unit. a statistical processing determination unit that determines the second operating state based on the converted data. Then, the driving state determination unit determines the driving state determination result by comparing the determination result by the transition condition determination unit and the determination result by the statistical processing determination unit. In this way, the driving state determination unit compares not only the first driving state determination result using the transition condition, but also the second driving result using the statistical processing, thereby improving accuracy. It becomes possible to determine the operating state. From the above, it is possible to accurately determine the transition of the driving state of the industrial vehicle.

The driving state determination unit may create a histogram based on the operation information detected by the detection unit, and determine the transition of the driving state based on the histogram. In this case, the operating state determination unit can further improve the accuracy by using the histogram in a state where the accuracy of the state transition is high.

The statistical processing determination unit may use a Gaussian mixture modeling method as the statistical processing. In this case, the statistical processing determination unit can improve the accuracy of the driving state determination by using the eigenvector.

The statistical processing determination unit may set the clustering setting number to a value larger than the transition number of the state transition model in the statistical processing by the Gaussian mixture modeling method. For example, if the number of clustering settings is too small, the operating state determination unit may find it difficult to determine the timing of feedback of the determination result of the second operating state. You can give feedback in a timely manner.

The data input to the driving state determination unit may include at least information for moving the vehicle and cargo handling work information by the cargo handling device. In this case, the driving state determination unit can determine the driving state in consideration of both the motion of the industrial vehicle and the motion of the cargo handling device.

An industrial vehicle according to an aspect of the present invention includes the industrial vehicle control device described above. This industrial vehicle can obtain the same actions and effects as those of the control device described above.

A control program for an industrial vehicle according to an aspect of the present invention causes the above-described control device for an industrial vehicle to function as a driving state determination unit. This industrial vehicle control program can obtain the same functions and effects as those of the control device described above.

ADVANTAGE OF THE INVENTION According to this invention, the control apparatus of an industrial vehicle, an industrial vehicle, and the control program of an industrial vehicle which can determine the transition of the driving|running state of an industrial vehicle accurately can be provided.

1 is a side view of a reach-type forklift equipped with a control device according to an embodiment of the present invention; FIG. FIG. 2 is a block configuration diagram showing the control device shown in FIG. 1 and components related thereto; (a) is a table showing each operating state in the state transition model, and (b) is a conceptual diagram showing the behavior of the forklift in the state transition model. It is a figure which shows a state transition model. It is a figure which shows the processing image of a driving|running state determination part. It is a graph which shows the determination result of the 2nd driving|running state based on statistical processing. (a) is a graph showing determination results of a first operating state, and (b) is a graph showing determination results of an operating state based on determination results of a first operating state and a second operating state. (a) is a graph showing the output of the detected signal, and (b) is a histogram corresponding to the signal in FIG. 8(a). 4 is a flow chart showing processing contents of a control device;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a side view of a reach-type forklift equipped with a control device according to this embodiment. In the following description, "left" and "right" will be used, but these correspond to "left" and "right" when viewed from the rear to the front. As shown in FIG. 1, a reach-type forklift truck (hereinafter simply referred to as a forklift truck) 50 as an industrial vehicle is a three-wheel vehicle type with two front driven wheels and one rear wheel drive. It is a battery car that runs on the power source of the housed battery. A forklift 50 includes a vehicle 2 and a cargo handling device 3 .

The vehicle 2 includes a pair of left and right reach legs 4 extending forward. The left and right front wheels 5 are rotatably supported by the left and right reach legs 4, respectively. The rear wheel 6 is one rear wheel and is a drive wheel that also serves as a steering wheel. A rear portion of the vehicle 2 is a standing type driver's seat 12 . An instrument panel 9 on the front side of the driver's seat 12 is provided with a cargo handling lever 10 for cargo handling operations and an accelerator lever 11 for forward and backward operations. A steering wheel 13 is provided on the upper surface of the instrument panel 9 .

The cargo handling device 3 is provided on the front side of the vehicle 2 . When the reach lever of the cargo handling lever 10 is operated, the reach cylinder 22 (see FIG. 2) is driven to extend and contract, so that the cargo handling device 3 moves forward and backward along the reach leg 4 within a predetermined stroke range. The cargo handling device 3 also includes a two-stage mast 23 , a lift cylinder 24 , a tilt cylinder 28 (see FIG. 2 ), and a fork 25 . When the lift lever of the cargo handling lever 10 is operated, the lift cylinder 24 is driven to extend and retract so that the mast 23 slides and retracts vertically, and the fork 25 moves up and down in conjunction with this.

Next, with reference to FIG. 2, the control device 100 of the forklift 50 according to this embodiment will be described in further detail. FIG. 2 is a block configuration diagram showing the control device 100 according to the present embodiment and components related thereto. As shown in FIG. 2, the control device 100 is connected to the steering wheel 13, the accelerator lever 11, and the cargo handling lever 10, and receives these driving commands. The cargo handling lever 10 includes a lift lever 10a, a reach lever 10b, and a tilt lever 10c. The lift lever 10a is a lever for operating a lift operation for moving the fork 25 in the vertical direction. The reach lever 10b is a lever for operating a reach operation for moving the cargo handling device 3 in the front-rear direction. The tilt lever 10c is a lever for operating a tilt operation for tilting the cargo handling device 3 .

The control device 100 is connected to the cargo handling drive system 40 and the traveling drive system 45, and transmits control signals to them. The cargo handling drive system 40 is a drive system that generates a driving force for operating the cargo handling device 3 . The traveling drive system 45 is a drive system that generates driving force for causing the vehicle 2 to travel.

The cargo handling drive system 40 drives each cylinder 22 , 24 , 28 by supplying hydraulic oil to the reach cylinder 22 , the lift cylinder 24 , and the tilt cylinder 28 . The cargo handling drive system 40 includes a cargo handling motor 41 and a valve portion 42 . The cargo handling motor 41 is a motor for driving a pump (not shown) that pressure-feeds hydraulic oil to each of the cylinders 22 , 24 , 28 . The valve unit 42 is a group of valves for controlling the supply destination and supply amount of hydraulic oil supplied to each cylinder 22 , 24 , 28 . For example, when the driver operates the lift lever 10a, the cargo handling motor 41 rotates at a rotation speed corresponding to the operation amount, and the valve unit 42 opens the valve corresponding to the lift cylinder 24 at an opening degree corresponding to the operation amount. open. At this time, the valve unit 42 keeps the valves corresponding to the reach cylinder 22 and the tilt cylinder 28 closed.

The travel drive system 45 includes a travel motor 46 . The travel motor 46 transmits torque to the rear wheels 6 (see FIG. 1). When the driver operates the accelerator lever 11 , the travel motor 46 rotates in a rotation direction based on the operating direction of the accelerator lever 11 at a rotation speed corresponding to the accelerator opening of the accelerator lever 11 . The travel drive system 45 also includes a steering mechanism that steers the rear wheels 6 based on the operation of the steering wheel 13 .

The control device 100 is a device for controlling the operation of the forklift 50 . The control device 100 controls the cargo handling drive system 40 and the traveling drive system 45 based on the operations of the respective operation units 10 , 11 , 13 . Further, the control device 100 according to this embodiment can determine the operating state of the forklift 50 . That is, the control device 100 can determine in real time what operating state the forklift 50 is in while the forklift 50 is in operation. The operating state is a work state defined by classifying cargo handling work performed by the driver of the forklift 50 based on the operation of the forklift 50 . The control device 100 determines the operating state based on a state transition model in which the operating states of the forklift 50 are classified. Here, before describing the specific configuration of the control device 100, a state transition model will be described with reference to FIGS. 3 and 4. FIG.

FIG. 3(a) is a table showing each driving state in the state transition model. FIG. 3(b) is a conceptual diagram showing the behavior of the forklift 50 in the state transition model. As shown in FIG. 3, the state transition model classifies a series of operations of the forklift 50 into 13 operating states S1 to S13. The operating state S1 is a state in which the vehicle is stopped without any cargo. The operating state S2 is a state in which the vehicle starts moving forward without any cargo. Driving state S3 is a state in which the vehicle turns forward with no load. The operating state S4 is a state in which the vehicle approaches the shelf SF for loading. The operating state S5 is a state in which cargo is loaded. The driving state S6 is a state in which the vehicle moves backward with cargo. The driving state S7 is a state in which the vehicle turns backward with cargo. The operating state S8 is a state in which the vehicle is stopped with cargo present. The operating state S9 is a state in which forward movement is started with cargo present. The driving state S10 is a state in which the vehicle is turning forward with cargo. The operating state S11 is a state in which the vehicle approaches the shelf SF for unloading. The operating state S12 is a state of unloading. The driving state S13 is a state in which the vehicle moves backward without any cargo.

FIG. 4 shows a state transition model created based on the above classification. The dashed arrows in this state transition model indicate the transition of the operation state of the forklift 50 . This state transition model can sufficiently represent the complexity of the operation pattern of the forklift 50 driver. The state transition model has self-loops and feedback/feedforward transition flows indicated by solid-line arrows in contrast to the basic model indicated by dashed-line arrows. The flow of solid-line arrows is for changing the state when various conditions do not match the characteristic quantity corresponding to the predetermined operating state. The feedback from the operating states S8, S9, S10, S11 to the operating state S7 assumes, for example, that the forklift 50 is repeatedly moved back and forth.

Next, referring to FIG. 2 again, a specific configuration of the control device 100 will be described. The control device 100 includes an ECU (Electronic Control Unit) that manages the device in an integrated manner. The ECU is an electronic control unit having a CPU [Central Processing Unit], a ROM [Read Only Memory], a RAM [Random Access Memory], a CAN [Controller Area Network] communication circuit, and the like. In the ECU, for example, programs stored in the ROM are loaded into the RAM, and the programs loaded into the RAM are executed by the CPU, thereby realizing various functions. The ECU may consist of a plurality of electronic units.

As shown in FIG. 2 , the control device 100 includes an operating state determination section 31 , a detection section 32 , a setting section 33 , a learning section 34 and a storage section 35 . A control program is installed in the control device 100 . This control program causes the control device 100 to function as an operating state determination unit 31 , a detection unit 32 , a setting unit 33 , a learning unit 34 and a storage unit 35 .

The operating state determination unit 31 determines the operating state of the forklift 50 based on the state transition model described above. In other words, the operating state determination unit 31 determines which of the operating states S1 to S13 the operating state of the forklift 50 during operation is. Further details of the driving state determination unit 31 will be described later.

The detection unit 32 detects operation information based on operations of various operation means in the forklift 50 . The detection unit 32 detects the steering direction and steering amount of the steering wheel 13 , the accelerator opening degree of the accelerator lever 11 , and the operation direction and operation amount of the cargo handling lever 10 . Specifically, the detector 32 detects an accelerator signal (ac signal) and a brake signal (br signal) from the operation of the accelerator lever 11 . The detection unit 32 uses these signals as data to estimate the running state and the stopped state. The detection unit 32 detects a forward/reverse signal (fb signal) from the operation direction. Further, the detection unit 32 detects a left/right signal (lr signal) indicating the tire direction based on the operation of the steering wheel 13, and detects a tire position signal (tp signal) indicating the tire position. In this embodiment, the tire position indicates the inclination (tire angle) of the rear wheels 6 that are driving wheels. The detection unit 32 detects the operating oil pressure (opc signal indicating oil pressure change) to the lift cylinder 24 of the cargo handling device 3 based on the lever operation of the cargo handling lever 10 . In addition, the detection unit 32 detects information related to loading and/or unloading from the lever operation, that is, a vertical signal (luf signal) related to the lift operation, a backward tilt signal (tbf signal) related to the tilt operation, a reach-in/out signal related to the reach operation ( rio signal).

The setting unit 33 sets transition conditions for the operating state determination unit 31 to perform determination. For example, if the storage unit 35 stores data of a plurality of transition conditions corresponding to skill levels, the setting unit 33 can grasp the driver's skill level and set appropriate transition conditions.

The learning unit 34 learns driving characteristics of the driver. Specifically, when the operating state determining unit 31 creates a histogram as shown in FIG. 8B for determining the operating state, the learning unit 34 may learn the content of the histogram. The learning unit 34 corrects the existing reference histogram based on the features of the newly created histogram. As a result, the result of the created histogram is reflected in the reference histogram. Note that, when there are a plurality of histograms created for a specific driver, the learning unit 34 may comprehensively determine the characteristics of the plurality of past histograms and correct the reference histogram.

The storage unit 35 stores various information used within the control device 100 . The storage unit 35 stores transition conditions used by the driving state determination unit 31 for determination. The storage unit 35 may store a plurality of transition conditions corresponding to the habits and skill levels of a plurality of drivers. In addition, the storage unit 35 stores learning results learned by the learning unit 34 .

As shown in FIG. 5, the driving state determination unit 31 acquires operation information based on operations of various operation means from the detection unit 32 . The data input to the driving state determination unit 31 includes at least information for moving the vehicle 2 and cargo handling work information by the cargo handling device. A first operating condition is determined by comparing with . On the other hand, the driving state determination unit 31 determines the second driving state by performing statistical processing on the acquired operation information. Then, the driving state determination unit 31 determines the driving state determination result by comparing the determination result by the transition condition determination unit 31A and the determination result by the statistical processing determination unit 31B. Specifically, the driving state determination unit 31 has a transition condition determination unit 31A, a statistical processing determination unit 31B, and a comprehensive determination unit 31C.

31 A of transition condition determination parts are comparing the detection result by the detection part 32, and transition conditions, and determine a 1st driving|running state. The transition condition determination unit 31A acquires operation information based on operations of various operation means from the detection unit 32 . 31 A of transition condition determination parts determine the driving|running state of the forklift 50 based on the transition conditions preset with respect to the determination parameter selected from the detection result. A transition condition is a condition set to determine that the operating state has transitioned to another operating state. That is, when the transition condition is satisfied, the transition condition determination unit 31A determines that the operating state of the forklift 50 has transitioned from one operating state to another operating state. In this way, when the transition condition determination unit 31A determines that the forklift 50 has transitioned to a certain operating state, the operating state is set as the "first operating state" determined based on the transition condition.

A specific example of the transition condition will now be described. Note that the transition conditions described here are merely examples, and may be changed as appropriate. Further, in the following description, "vector V _ji (1): accelerator", "vector V _ji (2): tire angle", "vector V _ji (3): lift lever operation", "V _ji (4) : reach lever operation" and "vector V _ji (5): tilt lever operation". For each operation, the vector V _ji (x) is close to one. Note that the transition condition determination unit 31A performs singular value decomposition on the covariance matrix of the Gaussian distribution associated with the cluster number clustered by the Gaussian mixture modeling method based on the detection result of the detection unit 32. Compute the values of the vector V _ji (x).

The transition condition from the operating state S1 to the operating state S2 is a condition for determining to start moving forward from a stop. For example, an accelerator signal is detected (ac>0) and there is a reaction of vector V _ji (1). That is. The transition condition from the driving state S2 to the driving state _S3 is a condition for determining that the vehicle continues to move forward and the speed increases. It is that there is a reaction of _ji (2). The transition condition from the operating state S3 to the operating state S4 is a condition for determining that the _forklift 50 starts to turn left while moving forward. There is no reaction of _ji (4) and V _ji (5). The transition condition from the operating state S4 to the operating state S5 is a condition for determining that the forklift 50 approaches the shelf SF straight while moving forward and prepares for loading. , the tire angle is almost straight. After such motion, there is lift and reach response, even if there is tire angle response. Furthermore, the reaction using lift and tilt becomes a value, and the vectors V _ji (3), V _ji (4), V _ji (5) also become a certain value. Alternatively, if there is no accelerator reaction, the reaction using lift or tilt will have a certain value, and the vectors V _ji (3), V _ji (4), and V _ji (5) will also have certain values. As the transition condition, a condition that can determine these situations may be set.

The transition condition from the operating state S5 to the operating state S6 may be a condition for determining that the loading is completed and the vehicle is backed up to leave the shelf SF. As such a condition, a predetermined condition that can confirm that the forklift 50 has a load may be employed. The transition state from the operating state S6 to the operating state S7 may be a condition for determining that the forklift 50 moves backward while turning right. In the operating state S6, the forklift 50 picks up the cargo, uses the cargo handling lever (lift, reach, or tilt), or focuses only on the lift, unloads the cargo from the shelf SF, and moves backward while moving. make it possible. Then, the forklift 50 moves backward and moves away from the shelf, and the tire angle becomes a large positive value (turns to the right) with almost no use of the cargo handling lever. As the transition condition, a condition that can determine these situations may be set.

The transition condition from the operating state S7 to the operating state S8 may be a condition for determining that the vehicle should stop after moving in reverse. For example, the accelerator signal should be approximately 0 and the tire angle should be approximately straight. It suffices if a condition for determining is set. The transition condition from the operating state S8 to the operating state S9 may be a condition for determining that the vehicle is moving forward from a stop. Any condition can be used as long as it can determine the unused state. As the transition condition from the operating state S9 to the operating state _S10 , it is sufficient to set a condition under which it can be determined that the vehicle continues to move forward and the speed increases. (3), V _ji (4)) can be any condition as long as it can be determined that there is no reaction.

The transition condition from the operating state S10 to the operating state S11 may be a condition for determining that the forklift 50 turns to the right while moving forward. , and the conditions that can determine a state in which there is no reaction of the parameters (vectors V _ji (3), V _ji (4), V _ji (5)) relating to the cargo handling lever. The transition condition from the operating state S11 to the operating state S12 may be any condition that can determine the state in which the forklift 50 approaches the rack for unloading. Any condition can be used as long as it is possible to determine a state in which the lift operation is performed while the SF is substantially facing the SF, and the cargo handling lever (vectors V _ji (3), V _ji (4)) reacts.

The transition condition from the operating state S12 to the operating state S13 may be a condition for determining that the forklift 50 unloads the load and backs away from the shelf SF. As the conditions for the operating state S12, predetermined conditions up to a state in which the cargo would have been unloaded are set. For example, a forklift 50 prepares for unloading. First, the reaction of the accelerator signal seems to disappear, and the tire angle becomes almost straight. After such motion, there is a reaction of the cargo handling levers (vectors V _ji (3), V _ji (4), V _ji (5)). In addition, there is a reaction that raises and lowers the lift independently. In addition, there is a reaction of reaching in and out, or tilting the fork backwards. Then there is the reaction of lowering the lift and returning the reach. Vectors V _ji (3), V _ji (4), and V _ji (5) will have certain values even if there is tire angle response in a nearly stationary state. After these reactions, there is a back reaction. Or there are more reactions after lifting or unloading a reach. As the transition condition, it is sufficient to set a condition that can determine these states.

The condition for transitioning from the operating state S13 to the operating state S1 may be any condition for determining that the series of operations should be completed and stopped. For example, there is reaction of the cargo handling lever while the forklift 50 is almost stopped. After that, the lift is brought down from a nearly stopped state. As a transition condition, what is necessary is just to set such a state that can be determined.

Note that even if the transition condition determination unit 31A determines the first operating state once, the determination may be erroneous. In order to correct the determination based on such an error, the transition condition determining section 31A may perform determination based on the transition condition for feedback.

The statistical processing determination unit 31B performs statistical processing on the detection result by the detection unit 32, and determines the second driving state based on the modeled data. The statistical processing determination unit 31B uses a Gaussian mixture model (GMM) as statistical processing. A Gaussian mixture model represents a model in which multiple Gaussian distributions (normal distributions) overlap. When multiple Gaussian distributions are superimposed, a mixture Gaussian model formula such as the following formula (1) is obtained. For example, when signals x(t) of accelerator, tire angle, lift lever operation, reach lever operation, and tilt lever operation are obtained at a predetermined sampling frequency (for example, 10 Hz), x(t) is distributed to K Gaussian distributions. The total value is denoted p(x) to indicate how likely it is to belong. Here, "K" indicates the number of clustering settings. This clustering setting number is set to a value larger than the transition number of the state transition model. In this embodiment, the number of transitions is 13, so the clustering setting number is set to a value greater than 13, such as "20". As a result, the statistical processing determination unit 31B can acquire the operating state at each time as a cluster number based on the signal x(t) at each time.

The statistical processing determination unit 31B determines the operating state obtained from the clustering number as the second operating state. FIG. 6A shows the result of the statistical processing determination unit 31B determining the second driving state using the Gaussian mixture modeling method with the clustering setting number set to "20". In this manner, the statistical processing determination unit 31B can determine the second operating state by referring to the correspondence between the operating states S1 to S13 and the clustering numbers 1 to 20. FIG. Statistical processing is not limited to using the Gaussian mixture model. For example, a hidden Markov model (HMM) may be employed. FIG. 6(b) shows the result of the statistical processing determination unit 31B determining the second driving state using the hidden Markov model. However, when the Hidden Markov Model is used to determine the driving state, the meaning of the state related to the transition state is experimentally or empirically given.

The comprehensive determination unit 31C determines the driving state determination result by comparing the determination result by the transition condition determination unit 31A and the determination result by the statistical processing determination unit 31B. In other words, the comprehensive determination unit 31C determines the driving state determination result by considering the second driving state of the statistical processing determination unit 31B as well as based on the first driving state determined by the transition condition determination unit 31A. For example, FIG. 7(a) is a graph showing the operating state determination result determined only by the first operating state. On the other hand, FIG. 7(b) is a graph showing the operating state determination result determined based on the first operating state and the second operating state. At timing t1, the comprehensive determination unit 31C determines the transition from the operating state S7 to the operating state S9 based on the first operating state. However, at timing t2, the comprehensive determination unit 31C immediately considers the second operating state and determines that the clustering number acquired by the statistical processing determination unit 31B is not the operating state S9 but the operating state S7. . As a result, the comprehensive determination unit 31C recognizes that the current operating state S9 is erroneous, returns to the operating state S7 without maintaining the operating state S9, and maintains the operating state S7. judge.

In this embodiment, the driving state determination unit 31 creates a histogram based on the operation information, and determines the transition of the driving state based on the histogram. For example, the driving state determination unit 31 calculates the amount of change in the tire position between the first driving state and the second driving state with respect to the tire position when transitioning to the first driving state, and the calculation result A histogram is created based on and a transition from the first operating state to the second operating state is determined based on the histogram. The driving state determination unit 31 calculates the amount of change in the tire position signal (tp signal) during transition from the first driving state to the second driving state (see, for example, FIG. 8A). Further, the driving state determination unit 31 creates a histogram based on the number of values of the amount of change in the signal that changes with time t (see, for example, FIG. 8B). The signal variation values are divided into six categories and normalized so that the sum of all values equals one. The horizontal axis of FIG. 7B indicates the number of each category. The vertical axis of FIG. 8(b) indicates the probability of the value of the amount of change in the signal present in each category. Thus, the histogram shown in FIG. 8(b) can be treated as a probability distribution regarding six categories. The driving state determination unit 31 prepares in advance a reference histogram created in advance. Then, the operating state determination unit 31 performs transition determination by matching the histogram created during operation of the forklift 50 with the reference histogram. The driving state determination unit 31 calculates the matching degree between the histogram being created and the reference histogram, and determines that the driving state has changed when the matching degree exceeds a predetermined threshold value. Although the matching method is not particularly limited, the driving state determination unit 31 uses the Kullback-Leibler divergence (KL-divergence) method to match the created histogram with the reference histogram, thereby determining the transition. you can It should be noted that the method by which the driving state determination unit 31 uses the histogram to determine the driving state is not limited to the one described above, and can be changed as appropriate.

Next, processing contents of the control device 100 according to the present embodiment will be described with reference to FIG. FIG. 9 is a flow chart showing the processing contents of the control device 100. As shown in FIG.

As shown in FIG. 9, the setting unit 33 sets transition conditions (step S10). For example, when the storage unit 35 stores a plurality of transition conditions according to the driver, the setting unit 33 selects an appropriate transition condition according to the driver and sets the selected transition condition. For example, if the storage unit 35 stores transition conditions and reference histograms unique to the driver, the setting unit 33 sets the reference histograms for the driver. It should be noted that the setting unit 33 can grasp what kind of driver drives the vehicle through an input operation or the like during driving. Note that the processing of S10 is performed when the storage unit 35 stores a plurality of transition conditions and it is necessary to determine which transition condition to use. , S10 may be omitted. Moreover, S10 may be omitted when the transition condition set in advance is used as it is.

After that, the driving state determination unit 31 acquires determination parameters necessary for determining the driving state (step S20). Here, the driving state determination section 31 acquires the operation information detected by the detection section 32 .

The driving state determination unit 31 determines whether or not there is a transition of the driving state based on the determination parameter acquired in S20 (step S30). Here, the operating state determination unit 31 determines the first operating state by the transition condition determination unit 31A described above, determines the second operating state by the statistical processing determination unit 31B, and determines the operating condition by the comprehensive determination unit 31C. Confirm the judgment result. When it is determined in S30 that there is no transition in the operating state, the operating state determination unit 31 determines that the current operating state continues, and repeats the process from S20. On the other hand, when it is determined in S30 that there is a transition in the driving state, the driving state determination unit 31 determines that the driving state has changed to a new driving state (step S40). The operating state determination unit 31 determines whether or not the operation of the forklift 50 has ended based on whether or not the forklift 50 is in an operable state (step S50). When the driving state determination unit 31 determines that the driving has not ended in S50, the processing is repeated from S20. If it is determined in S50 that the operation has ended, the process shown in FIG. 9 ends.

Next, functions and effects of the control device 100, the forklift 50, and the control program according to the present embodiment will be described.

In the control device 100 for the forklift 50 according to the embodiment of the present invention, the operating state determination unit 31 determines the operating state of the forklift 50 based on a state transition model in which the operating states of the forklift 50 are classified. In this manner, the operating state determination unit 31 makes determination based on the state transition model, thereby facilitating determination of the operating state of the forklift 50 . Here, the present inventors found the following knowledge as a result of earnest research. That is, when operation information based on operation of various operation means in the forklift 50 is detected and statistical processing is performed on the detection results, the processing results can be used to improve the accuracy of determining the operating state. I found something. Therefore, the driving state determination unit 31 compares the detection result of the detection unit and the transition condition to perform statistical processing on the transition condition determination unit 31A that determines the first driving state and the detection result of the detection unit 32. and a statistical processing determination unit 31B that determines the second driving state based on modeled data. Then, the driving state determination unit 31 determines the driving state determination result by comparing the determination result by the transition condition determination unit 31A and the determination result by the statistical processing determination unit 31B. In this way, the driving state determination unit 31 compares not only the first driving state determination result using the transition condition, but also the second driving result using the statistical processing, thereby improving accuracy. It becomes possible to determine the driving state. As described above, the transition of the operation state of the forklift 50 can be accurately determined.

The driving state determination unit 31 may create a histogram based on the operation information detected by the detection unit 32 and determine the transition of the driving state based on the histogram. In this case, the driving state determination unit 31 can further improve the accuracy by using the histogram in a state where the accuracy of the state transition is high.

The statistical processing determination unit 31B may use the Gaussian mixture modeling method as the statistical processing. In this case, the statistical processing determination unit 31B can improve the accuracy of the driving state determination by using the eigenvector.

The statistical processing determination unit 31B may set the clustering setting number to a value larger than the number of transitions of the state transition model in the statistical processing using the Gaussian mixture modeling method. For example, if the clustering setting number is too small, the operating state determination unit 31 has difficulty in timing the feedback of the determination result of the second operating state. You can give feedback in a timely manner.

The data input to the driving state determination unit 31 may include at least information for moving the vehicle 2 and cargo handling work information by the cargo handling device 3 . In this case, the driving state determination unit 31 can determine the driving state in consideration of both the motion of the vehicle 2 and the motion of the cargo handling device 3 in the forklift 50 . For example, in the case of a counter vehicle, there is no reach-in/reach-out signal, so it is possible to use detectable data that can be used to determine cargo handling.

The forklift 50 according to this embodiment includes the control device 100 for the forklift 50 described above. This forklift 50 can obtain the same actions and effects as the control device 100 described above.

The control program for the forklift 50 according to the present embodiment causes the above-described control device 100 for the forklift 50 to function as the operating state determination unit 31 . The control program for the forklift 50 can obtain the same actions and effects as the control device 100 described above.

The invention is not limited to the embodiments described above.

For example, in the above-described embodiment, a reach-type forklift was exemplified as an industrial vehicle, but various forklifts such as a counterbalance-type forklift may be employed. Note that detection of the reach operation is omitted when the system is used in a counterbalance forklift. Moreover, although the case where the traveling drive system is provided with the traveling motor was illustrated, the case where the traveling drive system is provided with the engine may be sufficient.

Also, in the above-described embodiments, the determination parameters corresponding to each embodiment were exemplified, but other determination parameters may be adopted without departing from the gist of the present invention. Also, the determination conditions are merely examples, and may be changed as appropriate.

DESCRIPTION OF SYMBOLS 2... Vehicle, 3... Cargo-handling apparatus, 31... Driving state determination part, 31A... Transition condition determination part, 31B... Statistical processing determination part, 32... Detection part, 50... Forklift (industrial vehicle), 100... Control apparatus.

Claims (7)

A control device for an industrial vehicle comprising a vehicle and a cargo handling device,
a detection unit that detects operation information based on the operation of various operation means in the industrial vehicle;
a driving state determination unit that determines the driving state of the industrial vehicle based on a state transition model in which the driving state of the industrial vehicle is classified;
The driving state determination unit is
a transition condition determination unit that determines a first operating state by comparing a detection result of the detection unit and a transition condition;
a statistical processing determination unit that performs statistical processing on the detection result of the detection unit and determines a second driving state based on modeled data;
A control device for an industrial vehicle, which establishes a driving state determination result by comparing a determination result by the transition condition determination unit and a determination result by the statistical processing determination unit. 2. The control device according to claim 1, wherein said driving state determination unit creates a histogram based on said operation information detected by said detection unit, and determines transition of said driving state based on said histogram. 3. The industrial vehicle control device according to claim 1, wherein said statistical processing determination unit uses a Gaussian mixture modeling method as said statistical processing. 4. The industrial vehicle control device according to claim 3, wherein said statistical processing determination unit sets a clustering setting number to a value larger than the number of transitions of said state transition model in said statistical processing by said Gaussian mixture modeling method. . The control of the industrial vehicle according to any one of claims 1 to 4, wherein the data input to the driving state determination unit includes at least information for moving the vehicle and cargo handling work information by the cargo handling device. Device. An industrial vehicle comprising the industrial vehicle control device according to any one of claims 1 to 5. An industrial vehicle control program that causes the industrial vehicle control device according to any one of claims 1 to 5 to function as the driving state determination unit.

Images