A working machine, a method for controlling a working machine, and a control system.

Abstract

A work machine having a pair of left and right traveling devices is moved so that the control point of the work machine follows the target path. [Solution] The control device uses a travel model to determine the amount of operation for each of the left and right pair of travel devices so that the control point of the work machine moves along the target path. The travel model is a model that estimates the state of the vehicle body after a unit of time from the position of the work machine and the amount of operation for the left and right pair of travel devices.

Description

[Technical Field]

[0001] This disclosure relates to a work machine, a method for controlling a work machine, and a control system. [Background technology]

[0002] Patent Document 1 discloses technology relating to autonomous driving of a work machine having a work device such as a blade and a pair of left and right traveling devices such as crawlers. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-166303 [Overview of the project] [Problems that the invention aims to solve]

[0004] When automatically controlling a work machine, it is necessary for the cutting edge of the machine to move along a target path in order to perform construction along that path. However, the control points, such as the cutting edge of the machine, are located far from the ground contact center of the travel device, making it difficult to make the machine's control points move along the target path. The purpose of this disclosure is to provide a work machine having a pair of left and right traveling devices that can be driven so that the control points of the work machine follow a target path, a control method for the work machine, and a control system. [Means for solving the problem]

[0005] According to one aspect of the present invention, the work machine comprises a vehicle body, a work machine located in front of or behind the vehicle body, a pair of left and right traveling devices attached to the vehicle body whose rotational speeds are independently controlled, and a control device that determines the amount of operation for each of the left and right traveling devices to move the work machine along a target path, using a travel model that estimates the state of the vehicle body after a unit of time based on the position of the work machine and the amount of operation of the pair of left and right traveling devices. [Effects of the Invention]

[0006] According to the above embodiment, a work machine having a pair of left and right traveling devices can be made to travel so that the control point of the work machine follows the target path. [Brief explanation of the drawing]

[0007] [Figure 1] This is a side view of the work machine according to the first embodiment. [Figure 2] This is a schematic diagram showing the power system of a work machine according to the first embodiment. [Figure 3] This diagram shows the configuration of the measurement system and control device of the work machine according to the first embodiment. [Figure 4] This figure shows an example of a rotation command table according to the first embodiment. [Figure 5] This is a flowchart showing the automatic steering control according to the first embodiment. [Figure 6] This figure shows the relationship between the control point and the alternative point according to the first embodiment. [Figure 7] This figure shows an example of a simulation using the driving model according to the first embodiment. [Figure 8] This figure shows an example of control when a work machine is moving backward, relating to a comparative example. [Figure 9] This figure shows an example of control when a work machine is moving in reverse according to the first embodiment. [Figure 10] This figure shows examples of the trajectories of control points when the work machine according to the first embodiment is moving forward and backward. [Modes for carrying out the invention]

[0008] <First Embodiment> The embodiments will be described in detail below with reference to the drawings. Figure 1 is a side view of the work machine according to the first embodiment. The work machine 100 according to the first embodiment is, for example, a bulldozer. The work machine 100 according to other embodiments may be other skid steering vehicles such as a hydraulic excavator or a tracked transport vehicle. The work machine 100 comprises a body 110, a running gear 120, a work machine 130, and a driver's cab 140.

[0009] The running gear 120 is installed on the underside of the vehicle body 110. The running gear 120 comprises a pair of crawlers 121 and sprockets 122. The pair of crawlers 121 are installed on the left and right sides of the vehicle body 110, respectively, and operate independently of each other. The crawlers 121 rotate when driven by the sprockets 122, causing the work machine 100 to move.

[0010] The work machine 130 is used for excavating and transporting materials such as soil and sand. The work machine 130 comprises a lift frame 131, a blade 132, and a blade lift cylinder 133. The blade 132 is positioned in front of the vehicle body 110.

[0011] The base end of the lift frame 131 is attached to the side of the vehicle body 110 via a pin extending in the vehicle width direction. The tip of the lift frame 131 is attached to the back surface of the blade 132 via a ball joint. This supports the blade 132 so that it can move vertically relative to the vehicle body 110. A cutting edge is provided at the lower end of the blade 132. The blade lift cylinder 133 is a hydraulic cylinder. The base end of the blade lift cylinder 133 is attached to the side of the vehicle body 110. The tip of the blade lift cylinder 133 is attached to the lift frame 131. The blade lift cylinder 133 expands and contracts due to hydraulic fluid, driving the lift frame 131 and the blade 132 upward or downward.

[0012] The cab 140 is a space where the operator sits and operates the work machine 100. The cab 140 is located on top of the vehicle body 110. A console and control devices are installed inside the cab 140.

[0013] The console is fitted with a control panel, instruments, and switches. The operator can visually check the status of the work machine 100 from the console. The control device receives input from the operator regarding the operation of a pair of crawlers 121 and blades 132. The control device consists of levers and pedals.

[0014] 《Power system》 Figure 2 is a schematic diagram showing the power system of the work machine according to the first embodiment. The work machine 100 is equipped with an engine 210, a PTO 220 (Power Take Off), a pair of HSTs 230 (Hydro Static Transmissions), a hydraulic pump 250, and a proportional control valve 260.

[0015] Engine 210 is, for example, a diesel engine. The PTO 220 transmits a portion of the power from the engine 210 to the hydraulic pump 250. In other words, the PTO 220 distributes the power from the engine 210 to a pair of HSTs 230 and hydraulic pumps 250. The HST230 transmits the driving force input to the input shaft at a variable speed to the output shaft. The HST230 comprises a hydraulic pump driven by the rotation of the input shaft and a hydraulic motor that rotates the output shaft. The HST230 controls the rotational speed of the output shaft by controlling the discharge flow rate of the hydraulic pump. In other embodiments of the HST230, the rotational speed of the output shaft may be controlled by the capacity of the hydraulic motor instead of the discharge flow rate of the hydraulic pump. The input shaft of the HST230 is connected to the PTO220, and the output shaft is connected to the sprocket 122. In other words, the HST230 transmits the driving force of the engine 210 distributed by the PTO220 to the sprocket 122. The output shafts of the pair of HST230s are connected to the left sprocket 122 and the right sprocket 122, respectively. The hydraulic pump 250 is driven by the power from the engine 210. The hydraulic fluid discharged from the hydraulic pump 250 is supplied to the blade lift cylinder 133 via the proportional control valve 260. The proportional control valve 260 controls the flow rate of the hydraulic fluid discharged from the hydraulic pump 250. In addition to the proportional control valve 260, the hydraulic pump 250 may also supply hydraulic fluid to other destinations, such as a steering clutch (not shown).

[0016] Figure 3 shows the configuration of the measurement system and control device of the work machine 100 according to the first embodiment. The measurement system of the work machine 100 acquires vehicle data representing the state of the work machine 100. The work machine 100 includes an IMU 310, a GNSS sensor 320, and a rotation speed sensor 330.

[0017] The IMU310 measures acceleration and angular acceleration for each axis of the vehicle coordinate system, which is represented by the X-axis extending in the longitudinal direction of the vehicle, the Y-axis extending in the lateral direction of the vehicle, and the Z-axis extending in the vertical direction of the vehicle, with the origin being the center of contact with the ground of the running gear 120. The GNSS sensor 320 measures the position, speed, and direction of the vehicle 110 in the global coordinate system based on signals from GNSS satellites. The rotation speed sensor 330 measures the rotation speed of the left and right sprockets 122.

[0018] Control device The work machine 100 is equipped with a control device 400 for controlling the work machine 100. The control device 400 outputs control signals to the fuel injector of the engine 210, the HST 230, and the proportional control valve 260 according to the amount of operation of the control device by the operator. The control device 400 also measures the position of the work machine 100 based on the measurement data of the measurement system and displays it on the console. The control device 400 may autonomously control the power system based on the position of the work machine 100 measured based on the measurement data of the measurement system. The control device 400 according to the first embodiment has two control modes: a manual driving mode and an automatic driving mode. The manual driving mode is a control mode in which the driving and steering of the work machine 100 are controlled based on the amount of operation of the control device. The automatic driving mode is a control mode in which the control device 400 calculates the amount of operation and controls the driving and steering of the work machine 100. When the control device 400 controls the work machine 100 in automatic driving mode, the control device 400 determines the amount of operation from the construction path information representing the outline of the area to be worked on by the blade 132 and the position of the control point of the cutting edge of the blade 132 so that the control point coincides with the construction path. The control device 400 calculates the operation amount based on at least the direction of travel of the vehicle and the positional relationship between the construction route and the work machine 100 (the relationship between the position of the construction route and the position of the work machine 100, the direction of the construction route as seen from the work machine 100, etc.).

[0019] The control unit 400 is a computer equipped with a processor 410, main memory 430, storage 450, and interface 470.

[0020] The storage 450 is a tangible, non-temporary storage medium. Examples of the storage 450 include magnetic disks, magneto-optical disks, and semiconductor memory. The storage 450 may be an internal medium directly connected to the bus of the control device 400, or an external medium connected to the control device 400 via the interface 470 or a communication line. The storage 450 stores a program for controlling the work machine 100. The storage 450 stores map data 452 of the construction site. The map data 452 includes a construction route that represents the outline of the area to be constructed with the blade 132. The construction route is represented by a combination of straight lines and arcs. The work machine 100 can properly construct the construction site by traveling so that the right or left end of the cutting edge of the blade 132 follows the construction route.

[0021] Figure 4 shows an example of a turning command table 451 according to the first embodiment. The storage 450 stores a turning command table 451 that shows the relationship between the speed ratio of a pair of crawlers 121 and the turning radius, which has been created in advance through experiments, etc. The speed ratio of a pair of crawlers 121 is the ratio of the speed of the crawler closer to the turning center to the speed of the crawler further from the turning center. When the work machine 100 is moving in a straight line, the speed ratio of the pair of crawlers 121 is 1. In the turning command table 451, the larger the turning radius, the larger the speed ratio of the pair of crawlers 121 becomes, approaching 1, and the smaller the turning curvature, the closer the speed ratio of the pair of crawlers 121 becomes, approaching 0. In the turning command table 451, the minimum value of the speed ratio of the crawlers 121 may be 0 or -1. When the speed ratio of the pair of crawlers 121 is 0, the crawler closer to the turning center stops. When the speed ratio of a pair of crawlers 121 is negative, the crawler closer to the turning center rotates in the opposite direction to the crawler further from the turning center. The turning command table 451 may be provided for each speed stage.

[0022] In other embodiments, the control device 400 may include, in addition to or instead of the above configuration, a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device). Examples of PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, some or all of the functions realized by the processor 410 may be realized by the integrated circuit.

[0023] The processor 410, by executing a program, includes an instruction input unit 411, a measurement data acquisition unit 412, a model prediction control unit 413, a command generation unit 414, and an output unit 415. Measurement data from the measurement system is input to the processor 410 via the interface 470.

[0024] The instruction input unit 411 receives operation signals for the crawler 121 and blade 132 from the control device. The instruction input unit 411 also receives automatic steering instructions from the console. The measurement data acquisition unit 412 acquires measurement data from the IMU 310, GNSS sensor 320, and rotation speed sensor 330. From the acquired measurement data, the measurement data acquisition unit 412 determines vehicle parameters such as the position of the tip of the blade 132, the speed of the running gear 120, and the turning curvature. Specifically, based on the position and speed of the antenna measured by the GNSS sensor 320, the angular velocity and acceleration of the vehicle body 110 measured by the IMU 310, and the known dimensions of the vehicle body 110, the measurement data acquisition unit 412 can calculate the position of the tip of the blade 132, the longitudinal speed and lateral speed of the running gear 120. The measurement data acquisition unit 412 can determine the turning curvature of the running gear 120 by dividing the turning angular velocity of the running gear 120 measured by the longitudinal speed of the running gear 120. In addition, in other embodiments, the measurement data acquisition unit 412 may calculate the speed of the travel device 120 by taking the average of the measured rotational speeds of the left and right sprockets 122. Alternatively, in other embodiments, the measurement data acquisition unit 412 may calculate the turning curvature of the travel device 120 from the ratio of the rotational speeds of the left and right sprockets 122 and the turning command table 451.

[0025] When performing automatic steering, the model prediction control unit 413 determines the turning curvature (target turning curvature) for the tip of the blade 132 to move along the construction path represented by the map data 452, based on the map data 452, the current position and direction indicated by the measurement data of the GNSS sensor 320, and the turning curvature of the travel device 120. The model prediction control unit 413 simulates the travel of the work machine 100 using the travel model of the work machine 100 and performs optimization processing to obtain a travel path that minimizes the discrepancy between the tip of the blade 132 and the construction path.

[0026] The command generation unit 414 uses the turning command table 451 to identify the speed ratio corresponding to the target turning curvature determined by the model prediction control unit 413. Based on the identified speed ratio, the command generation unit 414 generates pump capacity commands for a pair of HST230s. The command generation unit 414 records the identified speed ratio in the main memory 430, associating it with time. The output unit 415 outputs the pump capacity command generated by the command generation unit 414 to the HST 230.

[0027] 《Regarding the Travel Model》 The travel model used for model predictive control is the state Z of the work machine 100 at time (i×Δt) i and the turning curvature operation amount u i and is a model for obtaining the state Z of the work machine 100 after one unit of time i+1 . The travel model is represented by the following formula (1).

[0028]

Number

[0029] The state Z of the work machine 100 i is a vector having as elements the position Z of the control point xi of the work machine 100 at time (i×Δt), Z yi , the azimuth Z θi in which the work machine 100 faces, and the turning curvature Z cvi . That is, the state Z i is represented by the following formula (2).

[0030]

Number

[0031] The coefficient A, coefficient B, and term w of formula (1) i are values determined from the initial state Z0 of the work machine 100 and the construction path of the map data 452. The coefficient A is represented by formula (3). The coefficient B is represented by formula (4). The term w i is represented by formula (5).

[0032]

Number

Number

Number

[0033] Time Δt is a fixed time interval short enough that the trigonometric functions can be approximated by a straight line. State Z in equation (5) d i This represents an approximate value of the state at time (i × Δt), and u d i This represents the approximate value of the rotational curvature manipulation amount at time (i × Δt). Below, Z in the first embodiment d i The calculation method will be explained. First, the control device 400 calculates the approximate value Z of the initial state Z0 at time 0. d The components of 0 are determined as follows: The control device 400 determines Z d x0 (approximate value Z) d The x-component of 0 is set to the x-component of the point closest to the control point in the construction route of the map data 452. The control device 400 controls Z d y0 (approximate value Z) d The y component of 0 is set to the y component of the point closest to the control point in the construction route of the map data 452. The control device 400 controls Z d θ0 (approximate value Z) d The θ component of 0 is set to the tangential direction of the point in the construction route of the map data 452 that is closest to the ground contact center of the traveling device 120. The control device 400 is Z d cv0 (approximate value Z) d The cv component of 0 is set to the curvature of the path of the point in the construction route of the map data 452 that is closest to the ground contact center of the traveling device 120. The control device 400 is in the initial state Z d When 0 is determined, the x-component, y-component, tangential direction, and curvature of the point located a certain distance (M × V × Δt) along the construction path from the point referenced at that time are approximated by the state at time 1, Z. d Let's set it to 1. Similarly, the control device 400 obtains an approximate value of the state at any given time (i × Δt) by incrementing the time, Z d i It is possible to make a decision. Note that the coefficient A in equation (1) is obtained by adding Z to Z in equation (3). d i By substituting this, we can obtain the coefficient B, where u is in equation (4). d i This can be obtained by substituting the values. Thus, the control device 400 according to the first embodiment determines an approximate value related to the position of the control point based on the point in the construction path closest to the control point, and determines an approximate value related to the orientation and turning curvature of the work machine 100 based on the point in the construction path closest to the ground contact center. When the work machine 100 turns, the ground contact center of the traveling device 120 becomes the point of contact of the turning trajectory. Therefore, by determining the approximate values ​​related to the orientation and turning curvature based on the point in the construction path closest to the ground contact center rather than the point in the construction path closest to the control point, the control device 400 can appropriately control the turning operation. In the first embodiment, by giving the function g in the form of equation (6), we can obtain u from equations (4) and (5). d i Since it is canceled out, u d i This can be any value. In addition, the control device 400 according to other embodiments does not use the construction route of the map data 452, but instead determines the state at an arbitrary time (i × Δt) from the initial state Z0 using the differential model shown in equation (6), and sets it to state Z d i It may be used as follows. In this case, the rotational curvature manipulation variable u given to the differential model is the rotational curvature manipulation variable u obtained in the previous calculation. i You may use this. Alternatively, you may set u=0.

[0034] By the way, as shown in equations (3) to (5), coefficient A, coefficient B and term w i This is expressed by the partial derivatives of functions f and g. Functions f and g are functions that constitute the differential model of the work machine 100 shown in equation (6) below. The differential model is a model that represents the relationship between the state Z of the work machine 100 at time (i × Δt), the turning curvature manipulated variable u, and the derivative value Z′ of the state Z of the work machine 100.

[0035]

number

[0036] The coefficient V is the travel speed of the work machine 100. Constant term V slip x , V slip y These are the x and y components of the lateral slip velocity of the work machine 100, which are determined from the inclination of the work machine 100. The lateral slip velocity is determined based on the velocity of the traveling device 120 in the left-right direction (Y-axis direction of the vehicle coordinate system). Offset x off and y off These are the x-component length and y-component length from the ground contact center of the traveling device 120 to the control point on the work machine 130, respectively. In other words, offset x off and y off These are the x and y coordinates of the control point in the vehicle body coordinate system with respect to the ground contact center of the work machine 100. The coefficient τ is the delay time constant of the running gear 120. Functions f and g are determined in advance to satisfy equation (6) above.

[0037] Automatic steering method When the operator of the work machine 100 operates the console and inputs a command to start automatic steering, the control device 400 starts automatic steering control. The control device 400 repeatedly executes the automatic steering control shown below at control cycles (for example, 100 milliseconds).

[0038] Figure 5 is a flowchart showing the automatic steering control according to the first embodiment. The measurement data acquisition unit 412 acquires measurement data from the IMU 310, GNSS sensor 320, and rotation speed sensor 330 (step S1). From the acquired measurement data, the measurement data acquisition unit 412 determines the position (Z) of the tip (control point) of the blade 132 as a vehicle parameter. x , Z y , x off , y off ), the direction Z of the traveling device 120 θ , rotational curvature Z cvand velocity V, as well as sideslip velocity (V slip x , V slip y ) calculate (step S2).

[0039] The model prediction control unit 413 calculates the delay time t of the automatic steering control from the current time based on the measurement data acquired in step S1 and the vehicle parameters calculated in step S2. w The state Z0 of the work machine 100 at a later time (initial time) is determined (step S3). Specifically, the model prediction control unit 413 uses the travel model shown in equation (1) or the differential model shown in equation (6) to determine the delay time t w Determine the vehicle parameters after the specified time interval.

[0040] The model prediction control unit 413 determines whether the position of the control point is located in front of the ground contact center in the direction of travel in the state Z0 identified in step S3 (step S4). If the position of the control point is located behind the ground contact center in the direction of travel (step S4: NO), the model prediction control unit 413 determines whether the control point is close to the construction route in the map data 452 (step S5). For example, the model prediction control unit 413 determines whether the position of the control point (Z 0x , Z 0y The distance between the construction route and the azimuth (Z) is shorter than a predetermined distance threshold, and the direction (Z) 0θ If the difference in orientation between the control point and the tangent to the construction route is smaller than a predetermined angle threshold, it is determined that the control point and the construction route in map data 452 are close together.

[0041] If it is determined that the control point is located behind the ground contact center in the direction of travel, and that the control point is not close to the construction route in the map data 452 (step S5: NO), the model prediction control unit 413 determines the position of the control point in state Z0 (Z 0x , Z 0y The value of ) is replaced with the position of an alternative point located in front of the ground contact center in the direction of travel (step S6). Figure 6 is a diagram showing the relationship between the control point and the alternative point according to the first embodiment. The alternative point is a point that is symmetric to the control point with respect to an axis that passes through the ground contact center and extends in the left-right direction of the work machine 100. If the control point is located ahead of the ground contact center in the direction of travel (Step S4: YES), or if it is determined that the control point is close to the construction route in the map data 452 (Step S5: YES), the model prediction control unit 413 determines the position of the control point in state Z0 (Z 0x , Z 0y Do not change the value of ). The reason for using an alternative point when the control point is located behind the center of ground contact in the direction of travel, and when it is determined that the control point is not close to the construction route in map data 452, will be explained later.

[0042] Figure 7 shows an example of a simulation using the driving model according to the first embodiment. The model prediction control unit 413 determines the number of steps M to change the turning curvature manipulation amount in the simulation of driving using the driving model (step S7). The number of steps M will be explained below. The computational complexity of the optimization calculation varies depending on the number of optimization points N. Since the automatic steering control of the work machine 100 needs to be completed within the control cycle, an upper limit is set on the number of optimization points N. On the other hand, when realizing automatic steering control, if the path length D that is the target of the optimization calculation is too short, the travel path may be locally optimized. Therefore, it is preferable that the path length D that is the target of the optimization is sufficiently long. The path length D should be at least longer than the length of one vehicle of the work machine 100. The path length may be, for example, the distance when the work machine 100 travels at its maximum speed for a time (N × Δt). If the optimization calculation is performed for each cycle Δt until the path length D is completed, the number of optimization points N will increase, and it may not be possible to complete the calculation within the control cycle. The model prediction control unit 413 according to the first embodiment prevents an increase in the number of optimization points N by not changing the turning curvature manipulation amount during the step time (M × Δt). For example, the model prediction control unit 413 controls the rotation curvature manipulation amount u from time (M × Δt × j) to time (M × Δt × j + (M-1) × Δt). M×j ~u M×j+(M-1) The same rotational curvature control amount u m j The simulation will be performed as follows: Turn curvature control amount u M×j ~uM×j+(M-1) is called the step curvature u m j as well. FIG. 7 shows an example of a simulation when the number of steps M = 3. The number of steps M is the quotient of the step time and the time Δt. More specifically, the model predictive control unit 413 obtains the number of steps M by rounding to an integer the value obtained by dividing the path length D by the product of the number of optimization points N, the vehicle speed V, and the period Δt in order to fix the path length D and the number of optimization points N. The rounding process may be any of rounding up, rounding down, or rounding to the nearest integer. Note that when the period Δt is different from the value obtained by dividing the path length D by the product of the number of steps M, the number of optimization points N, and the vehicle speed V, the model predictive control unit 413 may replace the period Δt with the value obtained by dividing the path length D by the product of the number of steps M, the number of optimization points N, and the vehicle speed V.

[0043] Next, the model predictive control unit 413 identifies the first target point Z ref1 0 at the initial time, which is the point on the construction path closest to the control point, based on the map data 452 in the storage 450 and the initial state Z0 of the working machine 100 obtained in step S3 (step S8). The model predictive control unit 413 identifies N points separated from the first target point Z ref1 0 by a distance (M × V × Δt) on the construction path as the first target points Z ref1 1 to Z ref1 N (step S9). As a result, the model predictive control unit 413 can obtain N + 1 first target points Z ref1 0 to Z ref1 N (step S9). As a result, the model predictive control unit 413 can obtain N + 1 first target points Z

[0044] The model predictive control unit 413 identifies the position of the ground contact center of the traveling device 120 at the initial time from the initial state Z0, and based on the map data 452 in the storage 450 and the position of the ground contact center, identifies the second target point Z ref2 0 at the initial time, which is the point on the construction path closest to the ground contact center (step S10). The model predictive control unit 413 identifies N points separated from the second target point Z ref2 0 by a distance (M × V × Δt) on the construction path as the second target points Z ref2 1 to Z ref2N Specify it as (step S11). As a result, the model predictive control unit 413 determines N + 1 second target points Z ref2 0 to Z ref2 N can be obtained.

[0045] The model predictive control unit 413 determines N + 1 first target points Z ref1 0 to Z ref1 N and the second target points Z ref2 0 to Z ref2 N Based on these, the model predictive control unit 413 determines N + 1 target states Z tgt 0 to Z tgt N in step S12. In the first embodiment, Z tgt xi (the x component of the target state Z tgt i is set to the x component of the first target point Z ref1 i , Z tgt yi is set to the y component of the first target point Z ref1 i , Z tgt θi (the θ component of the state Z tgt i is set to the θ component of the second target point Z ref2 i , Z tgt θi (the θ component of the state Z tgt i is set to the cv component of the second target point Z ref2 i .

[0046] Based on the initial state Z0 and the N + 1 path states Z tgt 0 to Z tgt N , the model predictive control unit 413 performs an optimization calculation to obtain N step curvatures u m 0 to u m N-1 such that the loss function shown in Equation (7) is minimized under Equation (1) (step S13). As the optimization calculation, for example, quadratic programming can be used.

[0047]

number

[0048] The command generation unit 414 generates N step curvatures u based on the rotation command table 451 obtained in step S13. m 0~u m N-1 Of these, the step curvature u at the initial time m The speed ratio to achieve 0 is determined (step S14). The command generation unit 414 generates pump capacity commands for the pair of HST230 based on the determined speed ratio (step S15). The output unit 415 outputs the pump capacity commands generated in step S15 to the HST230 (step S16). As a result, the pair of crawlers 121 rotate and the work machine 100 moves.

[0049] In the first embodiment, the control device 400 causes the work machine 100 to travel along the route indicated by the map data 452. At this time, the control device 400 may also automatically control the height of the blade 132 in accordance with the terrain indicated by the map data 452.

[0050] Action / Effect The control device 400 according to the first embodiment determines the amount of manipulation required for the control point of the work implement 130 to move along the target path using a travel model. The travel model determines the position (Z) of the control point of the work implement 130. ix , Z iy ), the direction Z that the work machine 100 is facing iθ , and the rotational curvature Z icv And the amount of rotational curvature control of the traveling device 120 u i Therefore, this is a model for estimating the state of the work machine 100 after a unit of time. Thus, in the first embodiment, the control device 400, by providing a control point on the work machine 130, can make the work machine 100, which has a pair of left and right crawlers 121, travel so that the control point of the work machine 130 follows the target path.

[0051] In the driving model according to the first embodiment, the position of the first target point is determined based on the first point on the target path closest to the control point (Z ref1 ix、 Z ref1 iy ) and the azimuth and curvature (Z) of the second target point determined with reference to the second point on the target path closest to the ground contact center of the traveling device 120. ref2 iθ、 Z ref2 icv ) are combined to approximate the state of the work machine 100 at each time point, and the turning curvature control amount u i The state of the work machine 100 at each time point is estimated when given the given conditions. Thus, according to the travel model of the first embodiment, approximate values ​​of the state related to direction and turning curvature are determined based on the point in the construction path closest to the ground contact center, rather than the point in the construction path closest to the control point. As a result, the control device 400 can appropriately control the turning operation.

[0052] In the first embodiment, the control device 400 controls the position of the control point (Z) at each step time (M × Δt) which is M times the unit time Δt. Mix、 Z Miy ) and the first target point (Z) extracted at intervals of distance (M × V × Δt) from the first point on the target path closest to the control point position. ref1 ix、 Z ref1 iy The difference between the position of the work machine 100 and the position of the work machine 100 is minimized, and the orientation and curvature (Z Mix、 Z Miy ) and the azimuth and curvature (Z) of the second target point extracted at intervals of distance (M × V × Δt) from the second point on the target path closest to the ground contact center of the traveling device 120. ref2 iθ、 Z ref2 icvThe control device performs optimization calculations for each step time (M × Δt) so as to minimize the difference with the specified value. The step time is longer the slower the travel speed V of the work machine. By performing optimization calculations for each step time (M × Δt), the control device 400 can ensure the path length D while preventing an increase in the number of optimization points N. As a result, the control device 400 can complete the calculations within the control cycle and prevent the control from falling into a local solution. In other embodiments, the control device 400 may perform optimization calculations at each period Δt, regardless of the step time.

[0053] In the control device 400 according to the first embodiment, when the control point is located behind the ground contact center of the pair of left and right crawlers 121 in the direction of travel, the control point is replaced with an alternative point located in front of the ground contact center in the direction of travel, and the manipulated amount is determined. In this way, the control device 400 according to the first embodiment can prevent meandering when the work machine 100 is moving backward. Figure 8 is a diagram showing an example of control when the work machine 100 is moving backward according to a comparative example. When the control point is located behind the ground contact center in the direction of travel, depending on the optimization calculation method and the length of the path D, it is possible that the machine will meander along the path as shown in Figure 8. In other words, when the control point is located in front of the direction of travel, the position of the control point (Z ix、 Z iy ) After approaching the construction route, azimuth Z iθ and curvature Z icv The first part follows the path, and then the vehicle body 110 follows the path. In contrast, when the control point is located behind in the direction of travel, the position of the control point (Z) is as shown in Figure 8. ix、 Z iy Even if the ) approaches the construction route, azimuth Z iθ and curvature Z icv Because it does not follow the path, meandering may occur.

[0054] Figure 9 shows an example of control of the work machine 100 when it is moving backward according to the first embodiment. According to the first embodiment, if the control point is located behind the ground contact center of the pair of left and right crawlers 121 in the direction of travel, and the control point is not close to the construction path, the position of the control point is replaced with an alternative point located in front of the ground contact center in the direction of travel. As a result, the control device 400 controls the position of the alternative point (Z) in the same way as when the work machine is moving forward. ix、 Z iy ) After approaching the construction route, azimuth Z iθ and curvature Z icv The crawler tracks will follow the path, and then the vehicle body 110 will follow the path. This allows the control device 400 of the first embodiment to prevent meandering when the work machine 100 is moving in reverse. Furthermore, when the control point and the construction path become close, the control device 400 can switch the control point while the vehicle body 110 is following the path by performing control based on the control point rather than an alternative point. However, this is not limited to other embodiments. For example, in other embodiments, if some meandering is permissible when moving in reverse, the process of replacing with an alternative point may not be performed. Also, in other embodiments of the control device 400, while using autonomous control other than model predictive control (e.g., Pure Pursuit or PID control), if the control point is located behind the ground contact center of the pair of left and right crawlers 121 in the direction of travel, the position of the control point may be replaced with an alternative point in front of the ground contact center in the direction of travel.

[0055] Furthermore, when the manipulated variable is determined by model predictive control as in the first embodiment, meandering does not occur even without using alternative points. On the other hand, when the work machine 100 travels so that the pivot center traces the same pivot curvature, the trajectory traced by the control point differs depending on the direction of travel of the work machine 100. Figure 10 shows examples of the trajectory of the control point when the work machine 100 is moving forward and backward according to the first embodiment. As shown in Figure 10, the trajectory of the control point optimized by model predictive control differs depending on whether the machine is approaching forward or backward. In particular, when moving backward, if the machine attempts to enter the path at an angle, the cutting edge moves away from the path in the process of creating the angle, so it is unlikely that an manipulated variable that enters the path at an angle will be calculated in the optimization calculation. Therefore, when moving backward without using alternative points, the work machine 100 does not enter the path at an angle, so the control point may approach the path more slowly. In contrast, according to the first embodiment, when moving in reverse, the position of the control point is replaced with an alternative point at the rear to determine the amount of operation. Since the trajectory of the alternative point is the same as the trajectory of the control point when moving forward, according to the first embodiment, the way in which the tool approaches the path does not change when moving forward or backward, and the cutting edge can be brought closer to the path quickly.

[0056] The travel model according to the first embodiment includes a constant term that includes a lateral slip speed corresponding to the terrain on which the work machine 100 is located. This allows the control device 400 to achieve travel along the construction route while canceling out lateral slip of the work machine 100 that occurs according to the terrain. According to the first embodiment, the lateral slip speed is determined based on the speed of the travel device 120 in the left-right direction (Y-axis direction of the vehicle coordinate system), but is not limited to this. In other embodiments, if slipperiness information (gradient, geology, etc.) is associated with the map data 452 in advance, the control device 400 may determine the lateral slip speed based on the current position measured by the GNSS sensor 320 and the map data 452.

[0057] <Other Embodiments> Although one embodiment has been described in detail above with reference to the drawings, the specific configuration is not limited to that described above, and various design changes are possible. In other embodiments, the order of the above-described processes may be changed as appropriate. Also, some processes may be executed in parallel. The control device 400 according to the above embodiment may be composed of a single computer, or the configuration of the control device 400 may be divided among multiple computers, and the multiple computers may cooperate with each other to function as the control device 400. In this case, some of the computers constituting the control device 400 may be mounted inside the work machine, while the other computers may be provided outside the work machine.

[0058] The pair of crawlers 121 of the work machine 100 according to the above embodiment are driven by a pair of HST230s, but are not limited to this. For example, the pair of crawlers 121 according to another embodiment may be driven by a pair of electric motors. In this case, the control device 400 can achieve the speed ratio of the pair of crawlers 121 by current commands that instruct the amount of current of the pair of electric motors. Alternatively, for example, the pair of crawlers 121 according to another embodiment may distribute the power of a single power source to the pair of crawlers 121 by a differential. In this case, the control device 400 can achieve the speed ratio of the pair of crawlers 121 by commands that instruct the rotation ratio of the differential. Alternatively, for example, the speed of the pair of crawlers 121 according to another embodiment may be controlled by controlling the flow rate of hydraulic fluid with a control valve. [Explanation of symbols]

[0059] 100…Work machine 110…Body 120…Running gear 121…Crawler 122…Sprocket 130…Work machine 131…Lift frame 132…Blade 133…Blade lift cylinder 140…Driver's cab 210…Engine 220…PTO 230…HST 250…Hydraulic pump 260…Proportional control valve 310…IMU 320…GNSS sensor 330…Rotation speed sensor 400…Control unit 410…Processor 411…Instruction input unit 412…Measurement data acquisition unit 413…Model prediction control unit 414…Command generation unit 415…Output unit 430…Main memory 450…Storage 451…Swing command table 452…Map data 470…Interface

Claims

1. The car body and, A work machine located in front of or behind the vehicle body, A pair of left and right running gears, mounted on the vehicle body, whose rotational speeds are controlled independently of each other, A control device that determines the amount of operation for each of the left and right pair of traveling devices to move the work machine along a target path, using a travel model that estimates the state of the vehicle body after a unit of time based on the position of the work machine and the amount of operation of the left and right pair of traveling devices, A work machine equipped with the following features.

2. The aforementioned driving model is a model that estimates the state of the vehicle body after a unit of time based on the position of the work implement, the direction the vehicle body is facing, the turning curvature of the vehicle body, and the amount of manipulation. The work machine according to claim 1.

3. The control device determines the manipulated amount using the position of a first point on the target path, which is determined from the position of the work machine, and the orientation and curvature of a second point on the target path, which is determined from the ground contact center of the traveling device. The work machine according to claim 1.

4. The aforementioned driving model has a coefficient or term calculated from the position of the first point and the orientation and curvature of the second point. The work machine according to claim 3.

5. The control device is Using the target path, the current position of the work machine, and the travel model, an optimization calculation is performed on the operation amount for each step time such that the difference between the position of the work machine at each step time and the position of the first point at each step time, which is reached by moving a distance determined from the point on the target path closest to the current position of the work machine, is minimized. The work machine according to claim 1.

6. The step time is longer the slower the travel speed of the work machine. The working machine according to claim 5.

7. The control device is When the work implement is located behind the ground contact center of the pair of left and right traveling devices in the direction of travel, the operating amount is determined using the position of an alternative point in front of the ground contact center in the direction of travel, instead of the position of the work implement in the traveling model. The work machine according to claim 1.

8. The aforementioned alternative point is a point that is symmetrical to the work machine with respect to an axis that passes through the center of grounding and extends in the left-right direction of the work machine. The work machine according to claim 7.

9. The control device is If the work implement is located behind the ground contact center of the pair of left and right traveling devices in the direction of travel, and the distance between the work implement and the target path exceeds a threshold, the operating amount is determined using the position of an alternative point in front of the ground contact center in the direction of travel, instead of the position of the work implement in the traveling model. The work machine according to claim 7.

10. The aforementioned travel model has a constant term that includes a lateral slip speed corresponding to the terrain in which the work machine is located. The work machine according to claim 1.

11. The car body and, A work machine located in front of or behind the vehicle body, A pair of left and right running gears, mounted on the vehicle body, whose rotational speeds are controlled independently of each other, A control method using a control device for controlling a work machine equipped with the following: The control device uses a travel model that estimates the state of the work machine after a unit of time based on the position of the work machine and the amount of operation of the pair of left and right travel devices to determine the amount of operation of each of the pair of left and right travel devices so that the work machine moves along a target path. Control method.

12. The aforementioned driving model is a model that estimates the state of the vehicle body after a unit of time based on the position of the work implement, the direction the vehicle body is facing, the turning curvature of the vehicle body, and the amount of manipulation. The control method according to claim 11.

13. The control device determines the manipulated amount using the position of a first point on the target path, which is determined from the position of the work machine, and the orientation and curvature of a second point on the target path, which is determined from the ground contact center of the traveling device. The control method according to claim 11.

14. The control device performs optimization calculations for the manipulated variable at each step time, using the target path, the current position of the work machine, and the travel model, such that the difference between the position of the work machine at each step time and the position of the first point at each step time, which is reached by moving a distance determined from the step time from the point on the target path closest to the current position of the work machine, is minimized. The control method according to claim 11.

15. The step time is longer the slower the travel speed of the work machine. The control method according to claim 14.

16. The control device is When the work implement is located behind the ground contact center of the pair of left and right traveling devices in the direction of travel, the operating amount is determined using the position of an alternative point in front of the ground contact center in the direction of travel, instead of the position of the work implement in the traveling model. The control method according to claim 11.

17. The aforementioned alternative point is a point that is symmetrical to the work machine with respect to an axis that passes through the center of grounding and extends in the left-right direction of the work machine. The control method according to claim 16.

18. The control device is If the work implement is located behind the ground contact center of the pair of left and right traveling devices in the direction of travel, and the distance between the work implement and the target path exceeds a threshold, the operating amount is determined using the position of an alternative point in front of the ground contact center in the direction of travel, instead of the position of the work implement in the traveling model. The control method according to claim 17.

19. The aforementioned travel model has a constant term that includes a lateral slip speed corresponding to the terrain in which the work machine is located. The control method according to claim 11.

20. The car body and, A work machine located in front of or behind the vehicle body, A pair of left and right running gears, mounted on the vehicle body, whose rotational speeds are controlled independently of each other, A control system provided on the outside of a work machine equipped with the following: The state of the vehicle body, including the position of the work machine, and the amount of operation of the pair of left and right travel devices are input, and a travel model is used to estimate the state of the work machine after a unit of time, thereby determining the amount of operation of each of the pair of left and right travel devices for the work machine to move along a target path. Control system.

Images