WO2021199662A1 - Dispositif de commande de mouvement, procédé de commande de mouvement et support d'enregistrement - Google Patents

Dispositif de commande de mouvement, procédé de commande de mouvement et support d'enregistrement Download PDF

Info

Publication number
WO2021199662A1
WO2021199662A1 PCT/JP2021/003947 JP2021003947W WO2021199662A1 WO 2021199662 A1 WO2021199662 A1 WO 2021199662A1 JP 2021003947 W JP2021003947 W JP 2021003947W WO 2021199662 A1 WO2021199662 A1 WO 2021199662A1
Authority
WO
WIPO (PCT)
Prior art keywords
motion
operation amount
manipulated variable
parameter value
moving body
Prior art date
Application number
PCT/JP2021/003947
Other languages
English (en)
Japanese (ja)
Inventor
夏彦 佐藤
Original Assignee
日本電気株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電気株式会社 filed Critical 日本電気株式会社
Publication of WO2021199662A1 publication Critical patent/WO2021199662A1/fr

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/02—Control of vehicle driving stability
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion

Definitions

  • the present invention relates to a motion control device, a motion control method, and a recording medium.
  • Patent Document 1 describes a towing vehicle for making the dynamic characteristics when a towing vehicle (tractor) pulls a towed vehicle (trailer) closer to the dynamic characteristics when a motorcycle is used.
  • the behavior control device is described.
  • This towing vehicle behavior control device predicts the dynamic characteristics when the towing vehicle is in the motorcycle state when the towing vehicle is connected to the towed vehicle.
  • the towing vehicle behavior control device changes the dynamic characteristics of the connected vehicle to the dynamic characteristics of the motorcycle.
  • the drive unit is controlled so as to approach.
  • the towing vehicle when the towing vehicle is a single vehicle and the towed vehicle is connected, such as dynamic characteristics that change depending on the environment such as weather, road pavement, and wind, and differences in dynamic characteristics depending on the condition of the vehicle such as the load capacity of luggage.
  • dynamic characteristics that change depending on the environment such as weather, road pavement, and wind
  • dynamic characteristics depending on the condition of the vehicle
  • An example of an object of the present invention is to provide a motion control device capable of solving the above-mentioned problems.
  • the motion control device has an operation amount acquisition means for acquiring an operation amount (operation input value) for a controlled object, and motion data indicating the motion state (state of movement) of the controlled object.
  • the parameter value of the motion data acquisition means for acquiring the motion data, and the parameter value of the motion prediction model for outputting the predictive motion data indicating the motion predicted to occur in the control object in response to the input of the manipulated variable to the control object. It includes a parameter value determining means that is determined based on the motion data, and an manipulated variable correcting means that corrects the manipulated variable acquired by the manipulated variable acquisition means based on the parameter value determined by the parameter value determining means.
  • the motion control method acquires the operation amount for the control target, acquires the motion data indicating the motion state of the control target, and responds to the input of the operation amount for the control target. This includes determining a parameter value of a motion prediction model that outputs motion data indicating the motion of the controlled object based on the motion data, and correcting the acquired operation amount based on the determined parameter value.
  • the recording medium acquires the manipulated variable for the controlled object, acquires the motion data indicating the motion of the controlled object, and responds to the input of the manipulated variable for the controlled object.
  • the parameter value of the motion prediction model that outputs the motion data indicating the motion of the controlled object is determined based on the motion data, and the acquired manipulated variable is corrected based on the determined parameter value.
  • the embodiment of the present invention it is possible to reduce the difference in the dynamic characteristics of the moving body in addition to the difference in the dynamic characteristics between when the towing vehicle is a single vehicle and when the towed vehicle is connected.
  • the moving body in the first embodiment may be a moving body having a passenger or a moving body operated by an operator outside the moving body. An operator outside the moving body is also called an observer.
  • FIG. 1 shows a configuration example of a moving body according to the first embodiment.
  • the moving body 1 in the first embodiment includes a motion control device 2 and an actuator 14.
  • the motion control device 2 is a communication between a controller 10 for operating the moving body 1, a memory 11 for holding data such as a program, a CPU (central processing unit) 12 operated by program control, and the device. It includes a device I / O (Input / Output) 13 that mediates (communication between each part of FIG. 1), an acceleration sensor 15, and an encoder 16.
  • I / O Input / Output
  • the moving body referred to here is, for example, an automobile, a railroad vehicle, an aircraft, a ship, an automatic guided vehicle, or the like, which can move and control the movement.
  • the controller 10 outputs a signal indicating the amount of operation with respect to the moving body 1.
  • the controller 10 may accept an operation by an operator and output an operation amount according to the performed operation.
  • the controller 10 is provided with a steering wheel, an accelerator pedal, a brake pedal, and the like, receives a driving operation by an operator who is a driver, and outputs a signal indicating the amount of operation in the performed operation. You may try to do it.
  • the controller 10 may calculate the operation amount by the automatic operation and output a signal indicating the calculated operation amount.
  • the actuator 14 drives the moving body 1 according to the amount of operation output by the motion control device 2.
  • the actuator 14 may include an engine, a steering device (steering mechanism), and a brake, and move the moving body 1 according to a signal from the controller 10.
  • the amount of operation output by the motion control device 2 can be regarded as a control command value for the actuator 14. Therefore, it can be said that the motion control device 2 controls the motion of the actuator 14 by outputting the operation amount to the actuator 14, thereby controlling the motion of the moving body 1.
  • the acceleration sensor 15 measures the acceleration of the moving body 1.
  • the acceleration sensor 15 may be configured by using a three-axis acceleration sensor provided on the moving body 1 and may measure the acceleration in each axis.
  • the encoder 16 measures the amount of movement of the moving body 1. For example, when the moving body 1 is an automobile, the encoder 16 may measure the amount of rotation of the tire (the number of times the tire has rotated) and convert it into the moving distance of the moving body 1.
  • FIG. 2 is a block diagram showing a functional configuration example of the moving body 1.
  • the moving body 1 includes a motion control device 2 and a driving unit 23.
  • the motion control device 2 includes an operation amount acquisition unit 20, a data holding unit 21, an operation amount correction unit 22, a parameter value determination unit 24, and a state acquisition unit 25.
  • the operation amount acquisition unit 20 corresponds to the controller 10
  • the drive unit 23 corresponds to the actuator 14
  • the data holding unit 21 corresponds to the memory 11
  • the acceleration sensor 15 and the encoder 16 correspond to the state acquisition unit 25.
  • the position, velocity, acceleration, angular velocity, and angle can be estimated from the amount of movement and acceleration by the encoder 16.
  • the position, velocity, and acceleration of the moving body 1 for each sampling time obtained by the acceleration sensor 15 and the encoder 16 and the data associated with the sampling time are referred to as motion data. Therefore, the motion data is data indicating the motion of the moving body 1.
  • the data obtained by combining the motion data and the data indicating the manipulated variable acquired by the manipulated variable acquisition unit 20 for each sampling time is referred to as operation response data.
  • the position, velocity, and acceleration of the moving body 1 at a certain time indicate the motion state of the moving body 1 at that time.
  • the exercise data is not only data indicating exercise as described above, but also data indicating an exercise state.
  • the device I / O 13 included in the mobile body 1 for communicating between the devices of FIG. 1 controls communication (data input / output) of each part of FIG.
  • the operation amount correction unit 22 and the parameter value determination unit 24 function by operating the CPU 12 by a program on the memory 11.
  • the operation amount acquisition unit 20 acquires the operation amount for the moving body 1.
  • the operation amount acquisition unit 20 corresponds to an example of the operation amount acquisition means.
  • the state acquisition unit 25 acquires exercise data.
  • the state acquisition unit 25 corresponds to an example of the exercise data acquisition means.
  • the parameter value determination unit 24 determines the parameter value of the motion prediction model based on the motion data.
  • the parameter value determination unit 24 corresponds to an example of a parameter value determination means.
  • the motion prediction model predicts and predicts the motion performed by the moving body 1 according to the manipulated variable under the conditions indicated by the parameter value according to the setting of the parameter value and the input of the manipulated variable to the moving body 1.
  • This is a model that outputs predicted motion data indicating motion.
  • the motion prediction model includes a relative time t, a set x (t) of state variables including position and / or velocity, a history of a set u (t) of input quantities (by operation) at t, and a set a of parameters. It may be a relational expression that can predict x by using the function f.
  • the motion data output by the motion prediction model is also referred to as predicted motion data.
  • a which is a continuous degree of freedom in the motion prediction model
  • the parameter set a which is a continuous degree of freedom in the motion prediction model
  • Examples of such a motion prediction model include an equation of motion including parameters and a machine learning model that predicts the state of the system after the step time ⁇ t from the current state of the system and the amount of manipulation received.
  • the operation amount correction unit 22 corrects the operation amount acquired by the operation amount acquisition unit 20 based on the parameter value determined by the parameter value determination unit 24.
  • the operation amount correction unit 22 corresponds to an example of the operation amount correction means.
  • the data holding unit 21 holds various data. In particular, the data holding unit 21 stores the operation response data. In addition, the data holding unit 21 stores the motion prediction model and the reference model. Further, the data holding unit 21 stores the parameter value (parameter value of the motion prediction model) determined by the parameter value determining unit 24 and the parameter value of the reference model.
  • the reference model is a specific motion prediction model that calculates and outputs the motion that is the target of the control of the moving body 1.
  • the parameter value in the motion prediction model may be fixed to a constant value. That is, the reference model may be configured as a model showing the movement of a moving body under certain conditions.
  • the drive unit 23 drives the moving body 1 by the actuator 14 (FIG. 1).
  • the memory 11 holds a program for executing the function of the parameter value determination unit 24 and the function of the operation amount correction unit 22.
  • the memory 11 holds the parameters of the reference model and the initial parameters of the motion prediction model in advance (by the start of the operation of the motion control device 2).
  • the motion prediction model used in the embodiment and the motion prediction model on which the reference model used is based do not have to be the same.
  • the motion control device 2 uses the motion prediction model of the first vehicle to be driven and the reference model of the second vehicle different from the first vehicle, and the first vehicle has the same feeling as when driving the second vehicle.
  • the operation amount for the first automobile may be corrected so that the vehicle can be driven.
  • the functional units of the motion control device 2 operate as follows. Data in which the operation amount acquired by the operation amount acquisition unit 20 when the moving body 1 is operated is associated with the position, speed, acceleration, and time information obtained by the state acquisition unit 25 (the above operation response data). ) Is held by the state acquisition unit 25 in the data holding unit 21. When the operation response data of the determined amount of data is accumulated, the operation of the parameter value determination unit 24 starts.
  • Step S100 is a processing step until the operation response data of the determined amount of data is accumulated.
  • the parameter value determination unit 24 waits for the operation response data of the determined amount of data to be accumulated without performing any processing related to the parameter.
  • the processing shifts to the reading of the motion data and the operation amount (step S101).
  • step S101 the parameter value determination unit 24 reads out the set of the motion data and the operation amount (the operation response data described above) stored in the memory 11. As a result, the parameter value determination unit 24 can use the operation response data.
  • step S102 the parameter value determination unit 24 updates the parameter value of the motion prediction model using the read data.
  • the motion prediction model is an equation of motion (one-dimensional linear ordinary differential equation) as shown in the following equation (1) will be described as an example.
  • x is the position of the moving body 1 (at time t)
  • m is the total weight of the moving body 1
  • d is the coefficient of resistance proportional to the speed
  • u (t) is the manipulated variable at time t
  • g (u) Is a function representing the force applied to the moving body 1 when the operation of the manipulated variable u is performed.
  • the prime symbol "'" indicates that it is the time derivative of the variable.
  • m and d correspond to the example of the parameters of the motion prediction model.
  • the function g representing the motion prediction model is defined as a function that reflects the values of these parameters.
  • g is expanded by an appropriate function, and the effect of the manipulated variable u on the equation of motion is described by parameters.
  • polynomial expansion of g gives equation (2).
  • i be a positive integer
  • a i u i indicates the term when g is polynomial expanded.
  • a i is a coefficient (constant)
  • u i indicates the manipulated variable u to the i-th power.
  • a i / m and d / m are parameters.
  • "/ m" indicates division by the total weight m of the moving body 1.
  • the parameter value determination unit 24 can determine the parameter values of d / m and a / m by using a parameter estimation method such as the least squares method. can.
  • a parameter estimation method such as the least squares method.
  • the motion prediction model may be composed of a more complicated method such as a machine learning method.
  • the parameter value determination unit 24 can sequentially update the parameter value by the gradient descent method or the like. When the update of the parameter value of the motion prediction model is completed, the parameter value determination unit 24 saves the updated parameter value of the prediction model in the data holding unit 21 in step S103.
  • step S104 the parameter value determination unit 24 determines the end condition.
  • the process returns to step S100.
  • the parameter value determination unit 24 repeats the processes of step S101 and subsequent steps as soon as new data is accumulated.
  • step S104: YES the parameter value determination unit 24 ends the process of FIG.
  • step S110 the operation amount correction unit 22 acquires the operation amount that the operation amount acquisition unit 20 has been operated by the operator.
  • the manipulated variable correction unit 22 calculates the corrected manipulated variable using the parameters of the motion prediction model held by the data holding unit 21 (initial parameters or those updated by the parameter value determining unit 24).
  • the motion prediction model is a linear ordinary differential equation as in the above equation (1)
  • the equation of motion of the reference model be Eq. (4).
  • m 0 indicates the total weight of the moving object for which the reference model calculates the motion.
  • the moving object whose motion is calculated by the reference model is called a reference moving body.
  • d 0 indicates the coefficient of resistance proportional to the velocity in the reference moving object.
  • f (u) is a function that expresses the force applied to the reference moving body when the operation of the manipulated variable u is performed.
  • Equation (4) is derived from the substituted equation.
  • equation (5) when equation (5) is substituted into u (t) of equation (1), it becomes equation (6).
  • Equation (7) can be obtained by modifying equation (6).
  • equation (8) Multiplying both sides of equation (7) by m 0 / m gives equation (8), and a differential equation of the same form as equation (4) is obtained.
  • the apparent dynamic characteristics shown by the motion prediction model can be made equal to the dynamic characteristics shown by the reference model by absorbing the difference in the parameter values between the motion prediction model and the reference model.
  • the apparent dynamic characteristic shown by the motion prediction model here is the dynamic characteristic indicated by the combination of the correction of the manipulated variable and the motion prediction model with respect to the manipulated variable before the correction.
  • the value of m and the value of d in the equation (5) correspond to the example of the parameter value of the motion prediction model.
  • the value of m 0 and the value of d 0 in the equation (5) correspond to the example of the parameter value of the reference model.
  • the manipulated variable correction unit 22 corrects the manipulated variable with respect to the moving body 1 as in the above equation (5) to reduce the motion of the moving body 1. It can be the same as the movement of the reference moving body with respect to the manipulated variable before correction. As a result, the operator can operate the moving body 1 with the same feeling as when operating the reference moving body.
  • the reference moving body in this case may be a moving body different from the moving body 1.
  • the reference moving body in this case may be the moving body 1 under conditions different from those at the time of executing the operation, such as when the load of the moving body 1 changes.
  • FIG. 5 shows an example of correction performed by the operation amount correction unit 22.
  • the horizontal axis of the graph in FIG. 5 indicates the amount of operation before correction.
  • the vertical axis shows the amount of operation after correction.
  • FIG. 5 shows a graph when the value of u is taken on the horizontal axis and the value of ⁇ (u / 1.5) is taken on the vertical axis.
  • a / m and d / m are treated as parameters, respectively.
  • the values are fixed as constants as described above.
  • the manipulated variable correction unit 22 prepares a plurality of corrected manipulated variable candidates and selects one of them. You may. For example, when the manipulated variable acquired by the manipulated variable 20 is u, the manipulated variable correction unit 22 calculates u ⁇ ⁇ u, u ⁇ 2 ⁇ u, ... as candidates for the corrected manipulated variable. May be good.
  • ⁇ u is a constant predetermined (or according to the value of u) as the difference in the correction amount.
  • the manipulated variable correction unit 22 inputs the manipulated variable u to the reference model as the state after a certain time when it is input to the motion prediction model among the manipulated variables of u ⁇ ⁇ u, u ⁇ 2 ⁇ u, ...
  • the state after a certain time and the one closest to the meaning of the evaluation index D may be adopted.
  • the evaluation index D the distance between positions, which is a normal distance, the square root of the weighted sum of squares of each component, and the like can be used.
  • the evaluation function k that measures the closeness to the set of motion data from the present when the manipulated variable u is input to the reference model to a certain time later may be adopted.
  • the evaluation function k a function can be used in which the value obtained by adding the above Ds for each time is used as the return value.
  • the evaluation index an evaluation index that includes not only the evaluation of the proximity of exercise but also the evaluation other than the proximity of exercise such as the small amount of operation or the small energy consumption may be used.
  • the manipulated variable correction unit 22 is responsible for each of the elements of u as a vector of
  • U as a vector is also called an manipulated variable vector.
  • the manipulated variable correction unit 22 is optimized to approach the behavior of the reference model. Control may be performed.
  • x b is a vector indicating the position of the moving body (reference moving body) in the reference model.
  • x b' is a vector indicating the velocity of the moving object in the reference model.
  • x t is a vector representing the position of the moving body (reference moving body) in the motion prediction model.
  • x t' is a vector representing the velocity of the moving object in the motion prediction model.
  • a b , B b , At t , and B t all represent a matrix for describing the equation of motion.
  • u b is a vector representing the manipulated variable in the reference model.
  • u t represents the manipulated variable in the motion prediction model.
  • the " T " in the appendix represents the transpose of a matrix or vector.
  • Q and R are constant matrices and function as hyperparameters for determining how to evaluate the motion of the moving body 1.
  • the designer of the motion control device 2 determines in advance the values of Q and R (for example, at the time of designing the motion control device 2) according to the dynamic characteristics required for the moving body 1.
  • a new state variable (variable representing the motion of the moving body) x ⁇ is defined as in Eq. (14).
  • the manipulated variable correction unit 22 can determine the manipulated variable u to input J of the equation (19) based on the equation (15) using a normal optimum control framework.
  • the operation amount correction unit 22 outputs the operation amount calculated in step S111 to the drive unit 23 to control the movement of the moving body 1.
  • the operation amount correction unit 22 holds the operation amount and time output to the drive unit 23 in the data holding unit 21.
  • the manipulated variable held here is read out by the parameter value determining unit 24 in step S101 of FIG. 3 and used for updating the next parameter value.
  • the effect obtained in the first embodiment will be described.
  • the parameter value determining unit 24 updates the motion prediction model of the current controlled object
  • the motion of the controlled object such as the load capacity of the controlled object (moving body) and the road surface condition changes from time to time.
  • the characteristics can be predicted.
  • the manipulated variable correction unit 22 can bring the dynamic characteristics of the controlled object closer to the dynamic characteristics of the reference model.
  • the functions of the parameter value determination unit 24 and the operation amount correction unit 22 are not effective only for a specific target, and this embodiment can be generally applied to a device having a drive unit.
  • the operation amount acquisition unit 20 acquires the operation amount for the moving body 1.
  • the state acquisition unit 25 acquires motion data indicating the motion state of the moving body 1.
  • the parameter value determination unit 24 determines the parameter value of the motion prediction model based on the motion data, which outputs the predicted motion data indicating the motion predicted to occur in the moving body 1 in response to the input of the operation amount to the moving body 1. decide.
  • the operation amount correction unit 22 corrects the operation amount acquired by the operation amount acquisition unit 20 based on the parameter value determined by the parameter value determination unit 24.
  • the motion control device 2 it is possible to predict the motion even when the motion prediction model representing the motion of the moving body 1 has parameters such as different environments and load capacity that change the value. Then, the operation amount correction unit 22 corrects the operation amount, so that the movement of the moving body 1 can be brought closer to the movement indicated by the reference model. According to the motion control device 2, the motion of the moving body 1 approaches the motion indicated by the reference model, so that the operator operates the moving body 1 with a certain feeling even if the environment or the load capacity changes. It is possible to reduce the difficulty of operation and the feeling of strangeness.
  • the motion control device 2 it is possible to reduce the difference in the dynamic characteristics of the moving body in addition to the difference in the dynamic characteristics between when the towing vehicle is a single vehicle and when the towed vehicle is connected. Further, the framework of determining the parameter value of the motion prediction model is generally applicable to a device having a drive unit. In this respect, according to the motion control device 2, the dynamic characteristics of various types of devices having a drive unit can be brought close to the dynamic characteristics of the reference model.
  • the manipulated variable correction unit 22 has the manipulated variable acquired by the manipulated variable acquisition unit 20 into a composite function using the inverse function of the function representing the motion prediction model, the function representing the reference model, and the parameter values of the motion prediction model. Is input to calculate the corrected operation amount.
  • the reference model is a model that receives an input of a manipulated variable and outputs motion data indicating a reference motion.
  • the operation amount correction process is obtained in the form of a function, and the operation amount can be corrected by inputting the operation amount to this function, and the corrected operation amount is calculated in real time.
  • the processing load of the operation amount correction unit 22 for correcting the operation amount is relatively light.
  • the operation amount correction unit 22 uses the motion prediction model to select one of a plurality of corrected operation amount candidates obtained by performing different corrections on the operation amount acquired by the operation amount acquisition unit 20.
  • the reference model is selected using an evaluation index including an evaluation of the proximity of the motion output according to the input of the candidate of the above and the motion output according to the input of the manipulated variable acquired by the manipulated variable acquisition unit 20. Adopted as the amount of operation after correction.
  • the motion control device 2 even if the correction amount cannot be uniquely calculated, such as when the inverse function of the function representing the motion prediction model cannot be calculated, or even if the correction amount cannot be uniquely calculated in real time, the movement is performed. The movement of the body 1 can be brought close to the movement indicated by the reference model.
  • the manipulated variable correction unit 22 calculates the partial differential of each element of the manipulated variable vector of the index value indicating the degree of difference between the motion prediction model and the reference model, and corrects based on the obtained partial differential. Calculate the value.
  • the manipulated variable correction unit 22 can determine the direction of correction (whether to increase or decrease each element of the manipulated variable) based on the result of partial differentiation, and the moving body 1 can be determined. Corrections can be made to bring the movement closer to the movement indicated by the reference model.
  • FIG. 6 shows a configuration example of the remote motion control device according to the second embodiment.
  • the mobile system 3 includes a remote control device 4 and a mobile 5.
  • the remote operation control device 4 is for visually recognizing the state of a memory 50 for holding data such as a program, a CPU 51 operated by program control, an accelerator 52 for performing an operation operation, a handle 53, a brake 54, and a moving body. It includes a monitor 55, a remote side device I / O 56 that controls communication of each device in the remote operation control device 4, and a remote side communication device 57 for communicating between the mobile body 5 and the remote operation control device 4.
  • the remote operation control device 4 corresponds to an example of an operation control device.
  • the mobile body 5 includes a mobile body-side communication device 58 for communicating with the mobile body 5 and the remote operation control device 4, a mobile body-side device I / O 59 for communicating with each device in the mobile body 5, and the mobile body 5. It includes an actuator 60 for driving, an acceleration sensor 61 for measuring the acceleration of the moving body 5, an encoder 62 for measuring the position and speed of the moving body, and a camera 63 for imaging to assist remote operation. Further, although not shown in the figure, there may be a plurality of moving bodies. In that case, the remote control side operates one moving body from among several moving bodies, and the operation amount acquisition unit 20 operates. The operation target is switched by.
  • FIG. 7 is a block diagram showing a functional configuration example of the remote control device 4.
  • the mobile system 3 includes a remote control device 4 and a mobile 5.
  • the remote operation control device 4 includes an operation amount acquisition unit 70, a data holding unit 71, a video display unit 72, an operation amount correction unit 73, a parameter value determination unit 74, a remote side transmission unit 75, and a remote side reception. Includes part 76.
  • the moving body 5 includes a moving body side receiving unit 77, a moving body side transmitting unit 78, a driving unit 79, a state acquisition unit 80, and an imaging unit 81.
  • the operation amount acquisition unit 70 includes the accelerator 52, the handle 53, and the brake 54 of FIG. 6, and accepts the operation of the operator.
  • the operation amount acquisition unit 70 corresponds to the operation amount acquisition unit 20 (FIG. 2). Further, the operation amount acquisition unit 70 corresponds to an example of the operation amount acquisition means.
  • the data holding unit 71 corresponds to the memory 50 and stores various data.
  • the data holding unit 71 corresponds to the data holding unit 21.
  • the operation amount correction unit 73 and the parameter value determination unit 74 both operate when the CPU 51 executes a program on the memory 50.
  • the parameter value determination unit 74 corresponds to the parameter value determination unit 24 and determines the parameter value of the motion prediction model based on the motion data.
  • the parameter value determination unit 74 corresponds to an example of the parameter value determination means.
  • the operation amount correction unit 73 corresponds to the operation amount correction unit 22, and corrects the operation amount acquired by the operation amount acquisition unit 70 based on the parameter value determined by the parameter value determination unit 74.
  • the operation amount correction unit 73 corresponds to an example of the operation amount correction means.
  • the image display unit 72 is configured by using the monitor 55, and displays an image around the moving body 5 taken by the camera 63.
  • the remote-side transmitting unit 75 and the remote-side receiving unit 76 are configured by using the remote-side communication device 57.
  • the moving body side transmitting unit 78 and the moving body side receiving unit 77 are configured by using the moving body side communication device 58.
  • Various data are transmitted and received between the remote side transmitting unit 75 and the moving body side receiving unit 77.
  • Various data are transmitted and received between the moving body side transmitting unit 78 and the remote side receiving unit 76.
  • the drive unit 79 is configured by using the actuator 60, and drives the moving body 5 based on the operation amount after correction by the operation amount correction unit 73.
  • the drive unit 79 corresponds to the drive unit 23.
  • the state acquisition unit 80 is configured by using the encoder 62 and the acceleration sensor 61.
  • the state acquisition unit 80 corresponds to the state acquisition unit 25 and acquires exercise data.
  • the state acquisition unit 80 corresponds to an example of a state acquisition means.
  • the image pickup unit 81 is configured by using the camera 63, and captures an image of the surroundings of the moving body 5. For example, the imaging unit 81 captures an image when the moving body 5 looks around the moving body 5.
  • the difference between the second embodiment and the first embodiment is that the operator performs the operation while viewing the image captured by the imaging unit 81 on the image display unit 72.
  • the transmission / reception of the image and the transmission / reception of the manipulated variable and the state variable are performed by the remote side transmitting unit 75 and the moving body side transmitting unit 78, and the remote side receiving unit 76 and the moving body side receiving unit 77.
  • the operation of the operation amount correction unit 73 and the parameter value determination unit 74 is the same as that of the first embodiment except that the state variable received from the moving body side is used.
  • the second embodiment is the same as the case of the first embodiment except for these points.
  • the remote control device 4 is used to communicate with the remote transmission unit 75 and the mobile reception unit 77, and the communication between the mobile transmission unit 78 and the remote reception unit 76. It is possible to remotely control 5.
  • the parameter value determination unit 74 updates the parameter value of the motion prediction model of the moving body 5 which is the current control target, the motion prediction representing the motion of the moving body 5 is performed. It is possible to predict motion even when the model has parameters that change values such as different environments and load capacity.
  • the dynamic characteristic of the moving body 5 to be controlled can be brought closer to the dynamic characteristic indicated by the reference model. For example, when the control target is switched between a plurality of vehicles, this effect becomes larger, and when the operation is switched to a vehicle having a different load capacity or a vehicle of a different type, the operator due to the difference in dynamic characteristics It is possible to reduce the difficulty of operation and the feeling of strangeness.
  • FIG. 8 is a diagram illustrating a configuration example of the remote operation system according to the third embodiment.
  • the remote operation system 6 in the third embodiment includes a computer 200, operating equipment 201, and communication equipment 202.
  • the computer 200 includes a part of the operation amount acquisition unit 70 of FIG. 7 (a part other than the accelerator 52, the handle 53, and the brake 54 (a part that performs processing)), a data holding unit 71, an image display unit 72, and the like.
  • the operation amount correction unit 73 is included.
  • the operation equipment 201 includes a part of the operation amount acquisition unit 70 (accelerator 52, handle 53, and brake 54) of FIG. 7.
  • Communication equipment 202 includes a remote side transmitting unit 75 and a remote side receiving unit 76.
  • the communication equipment 202 communicates with each of the first automobile 203, the second automobile 204, and the third automobile 205.
  • the first automobile 203, the second automobile 204, and the third automobile 205 each include a moving body side receiving unit 77, a moving body side transmitting unit 78, a driving unit 79, a state acquisition unit 80, and an imaging unit 81 in FIG. 7, respectively. I'm out.
  • the first vehicle 203, the second vehicle 204, and the third vehicle 205 are self-driving vehicles that are normally driven autonomously, and are monitoring / operators who may drive remotely. Monitors the operation (autonomous driving) of these vehicles. While not driving remotely, the first vehicle 203, the second vehicle 204, and the third vehicle 205 continue to send motion data to the data holding unit 71 in the computer 200, and the computer 200 functions as the parameter value determining unit 74. Is executed to update the parameter value continuously or repeatedly.
  • the computer 200 functions as the operation amount correction unit 73 for the first automobile 203 and is updated. Correct the operation amount using the parameter value. As a result, the dynamic characteristics of the first automobile 203 are corrected so as to be close to the dynamic characteristics shown by the reference model, and the monitoring / operator can drive relatively easily.
  • the monitoring / operator changes the operation target from the first vehicle 203 to the second vehicle 204 or the third vehicle 205 of a different vehicle type
  • the motion data of the second vehicle 204 and the third vehicle 205 are acquired and retained. Since the parameters have been updated, the computer 200 corrects the operation amount so that the dynamic characteristics of the vehicle to be operated have the dynamic characteristics close to the dynamic characteristics shown by the reference model.
  • the remote control system 6 the monitoring / operator can perform remote control without being aware of changes in dynamic characteristics even when switching the operation target, and in this respect, the effect of facilitating remote control is achieved. Obtainable.
  • FIG. 9 shows a configuration example of the motion control device according to the fourth embodiment.
  • the motion control device 300 includes an operation amount acquisition unit 301, an exercise data acquisition unit 302, a parameter value determination unit 303, and an operation amount correction unit 304.
  • the operation amount acquisition unit 301 acquires the operation amount for the moving body.
  • the motion data acquisition unit 302 acquires motion data indicating the motion state of the moving body.
  • the parameter value determination unit 303 determines the parameter value of the motion prediction model that outputs the predicted motion data indicating the motion predicted to occur in the moving body in response to the input of the operation amount to the moving body, based on the motion data. ..
  • the operation amount correction unit 304 corrects the operation amount acquired by the operation amount acquisition unit 301 based on the parameter value determined by the parameter value determination unit 303.
  • the motion control device 300 it is possible to predict the motion even when the motion prediction model representing the motion of the moving body has parameters such as different environments and load capacity that change the value. Then, the operation amount correction unit 304 corrects the operation amount, so that the movement of the moving body can be brought closer to the movement indicated by the reference model (the model showing the reference movement). According to the motion control device 300, the motion of the moving body approaches the motion indicated by the reference model, so that the operator can operate the moving body with a certain feeling even if the environment or the load capacity changes. , Difficulty of operation and discomfort can be reduced.
  • the motion control device 300 it is possible to reduce the difference in the dynamic characteristics of the moving body in addition to the difference in the dynamic characteristics between when the towing vehicle is a single vehicle and when the towed vehicle is connected. Further, the framework of determining the parameter value of the motion prediction model is generally applicable to a device having a drive unit. In this respect, according to the motion control device 300, the dynamic characteristics of various types of devices having a drive unit can be brought close to the dynamic characteristics of the reference model.
  • FIG. 10 shows an example of a procedure for processing the motion control method according to the fifth embodiment.
  • the process shown in FIG. 10 includes an operation amount acquisition step (step S301), an exercise data acquisition step (step S302), a parameter value determination step (step S303), and an operation amount correction step (step S304).
  • step S301 the operation amount for the moving body is acquired.
  • step S302 motion data indicating the motion state of the moving body is acquired.
  • step S303 the parameter value of the motion prediction model that outputs the motion of the moving body in response to the input of the manipulated variable for the moving body is determined based on the motion data.
  • step S304 the acquired operation amount is corrected based on the determined parameter value.
  • the process of FIG. 10 it is possible to predict the motion even when the motion prediction model representing the motion of the moving object has parameters such as different environments and load capacity that change the value. Then, by correcting the manipulated variable in the manipulated variable correction step, the motion of the moving body can be brought closer to the motion indicated by the reference model (model showing the reference motion). According to the process of FIG. 10, the movement of the moving body approaches the movement shown by the reference model, so that the operator can operate the moving body with a certain feeling even if the environment or the load capacity changes. , Difficulty of operation and discomfort can be reduced.
  • the framework of determining the parameter value of the motion prediction model is generally applicable to a device having a drive unit.
  • the dynamic characteristics of various types of devices having a drive unit can be brought close to the dynamic characteristics of the reference model.
  • control target is a moving body
  • control target is not limited to the moving body in any of the embodiments.
  • the control target may be a robot arm.
  • the robot arm in this case may be one that is automatically controlled or semi-automatically controlled, or one that is manually operated.
  • FIG. 11 is a schematic block diagram showing the configuration of a computer according to at least one embodiment.
  • the computer 700 includes a CPU (Central Processing Unit) 710, a main storage device 720, an auxiliary storage device 730, and an interface 740.
  • CPU Central Processing Unit
  • any one or more of the above-mentioned motion control device 2 and remote control device 4 may be mounted on the computer 700.
  • the operation of each of the above-mentioned processing units is stored in the auxiliary storage device 730 in the form of a program.
  • the CPU 710 reads the program from the auxiliary storage device 730, expands it to the main storage device 720, and executes the above processing according to the program. Further, the CPU 710 secures a storage area corresponding to each of the above-mentioned storage units in the main storage device 720 according to the program.
  • the operations of the operation amount correction unit 22 and the parameter value determination unit 24 are stored in the auxiliary storage device 730 in the form of a program.
  • the CPU 710 reads a program from the auxiliary storage device 730, deploys it to the main storage device 720, and executes the operation of each unit according to the program.
  • the CPU 710 secures a storage area corresponding to the data holding unit 21 in the main storage device 720.
  • Data acquisition by the operation amount acquisition unit 20 and the state acquisition unit 25 is executed by the interface 740 having a communication function and communicating with another device according to the control of the CPU 710.
  • the operations of the operation amount correction unit 22 and the parameter value determination unit 24 are stored in the auxiliary storage device 730 in the form of a program.
  • the CPU 710 reads a program from the auxiliary storage device 730, deploys it to the main storage device 720, and executes the operation of each unit according to the program.
  • the CPU 710 secures a storage area corresponding to the data holding unit 21 in the main storage device 720.
  • Data acquisition by the operation amount acquisition unit 20 and the state acquisition unit 25 is executed by the interface 740 having a communication function and communicating with another device according to the control of the CPU 710.
  • a program for realizing all or a part of the functions of the motion control device 2 and the remote operation control device 4 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system. The processing of each part may be performed by executing.
  • the term "computer system” as used herein includes hardware such as an OS (operating system) and peripheral devices.
  • "Computer readable recording medium” includes flexible disks, magneto-optical disks, portable media such as ROM (Read Only Memory) and CD-ROM (Compact Disc Read Only Memory), hard disks built into computer systems, and the like.
  • a storage device includes flexible disks, magneto-optical disks, portable media such as ROM (Read Only Memory) and CD-ROM (Compact Disc Read Only Memory), hard disks built into computer systems, and the like.
  • the above-mentioned program may be a program for realizing a part of the above-mentioned functions, and may be a program for realizing the above-mentioned
  • the present invention may be applied to a motion control device, a motion control method, and a recording medium.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Feedback Control In General (AREA)

Abstract

La présente invention concerne un dispositif de commande de mouvement comprenant un moyen d'acquisition de quantité de commande qui acquiert une quantité de commande par rapport à une cible de commande, un moyen d'acquisition de données de mouvement qui acquiert des données de mouvement indiquant un état de mouvement de la cible de commande, un moyen de détermination de valeur de paramètre qui détermine, sur la base des données de mouvement, une valeur de paramètre d'un modèle de prédiction de mouvement qui délivre en sortie des données de mouvement prédits indiquant un mouvement prédit pour se produire dans la cible de commande en réponse à l'entrée de la quantité de commande par rapport à la cible de commande, et un moyen de correction de quantité de commande qui corrige la quantité de commande acquise par le moyen d'acquisition de quantité de commande sur la base de la valeur de paramètre déterminée par le moyen de détermination de valeur de paramètre.
PCT/JP2021/003947 2020-03-31 2021-02-03 Dispositif de commande de mouvement, procédé de commande de mouvement et support d'enregistrement WO2021199662A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020065038 2020-03-31
JP2020-065038 2020-03-31

Publications (1)

Publication Number Publication Date
WO2021199662A1 true WO2021199662A1 (fr) 2021-10-07

Family

ID=77928071

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/003947 WO2021199662A1 (fr) 2020-03-31 2021-02-03 Dispositif de commande de mouvement, procédé de commande de mouvement et support d'enregistrement

Country Status (1)

Country Link
WO (1) WO2021199662A1 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006013922A1 (fr) * 2004-08-06 2006-02-09 Honda Motor Co., Ltd. Dispositif de contrôle pour véhicule

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006013922A1 (fr) * 2004-08-06 2006-02-09 Honda Motor Co., Ltd. Dispositif de contrôle pour véhicule

Similar Documents

Publication Publication Date Title
CN111142379B (zh) 一种自动驾驶车辆的动力学预测控制系统和方法
US11104352B2 (en) Vehicle control system
CN109421630B (zh) 用于监测自主车辆的健康的控制器架构
JP4400418B2 (ja) 車間距離制御装置及び車間距離制御方法並びに運転操作支援装置及び運転操作支援方法
US10929995B2 (en) Method and apparatus for predicting depth completion error-map for high-confidence dense point-cloud
CN111923927B (zh) 用于交互感知交通场景预测的方法和装置
JP2008290713A (ja) ビークル制御の改良
JP2005339241A (ja) モデル予測制御装置および車両用推奨操作量生成装置
CN110799402B (zh) 车辆控制装置
US11584342B2 (en) Real-time performance handling virtual tire sensor
JP2020527502A (ja) 車両の運動を制御する制御システム及び方法
CN108717268A (zh) 基于最优控制与安全距离的自动驾驶最速操纵控制系统及其控制方法
JP7039389B2 (ja) ロボットシステム及びロボット制御方法
JP7212702B2 (ja) 車両制御装置、車両制御方法、及び車両制御システム
CN113548032B (zh) 改善车辆控制系统的控制子系统的性能的方法和系统
CN112238857A (zh) 自动驾驶车辆的控制方法
CN115707609A (zh) 驾驶员命令预测
Guidolini et al. Neural-based model predictive control for tackling steering delays of autonomous cars
WO2021199662A1 (fr) Dispositif de commande de mouvement, procédé de commande de mouvement et support d'enregistrement
CN118270091A (zh) 用于自动车辆的转向控制系统和方法
KR20180116663A (ko) 자율주행 제어권 전환 시스템 및 방법
JP6250184B2 (ja) 車両情報制御装置
CN111738046A (zh) 对用于基于深度学习的装置的学习的虚拟世界模拟器的物理引擎进行标定的方法及装置
CN115081612A (zh) 用以改进机器人策略学习的设备和方法
CN112700001A (zh) 用于深度强化学习的认证对抗鲁棒性

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21778882

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21778882

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP