Tax characteristics selection system, tax characteristics selection procedure

Abstract

Control characteristic selection system for searching for a control characteristic of an actuator that contributes to a driving characteristic of a moving body, comprising a database in which a multitude of combinations of the control characteristic and a sensory evaluation rating are stored, a sensory evaluation rating unit for inputting the control characteristic as a current characteristic and outputting the sensory evaluation rating as a current evaluation rating, an input unit for receiving a target value which is the sensory evaluation rating for targeting, and a selection unit which is configured to select from the control characteristics stored in the database the control characteristic in which the sensory evaluation rating is similar to the target value and is similar to the current characteristic as a selected characteristic.

Description

Technical field

[0001] The present invention relates to a control characteristic selection system and a control characteristic selection method, State of the art

[0002] When an actuator is actuated according to a human input, it is essential to perform an adjustment taking human senses into account, that is, to adjust the control characteristics of the actuator. Patent document 1 discloses a control parameter adjustment teaching device for an electric power steering control device that performs control based on various types of control parameter data, wherein the control parameter adjustment teaching device comprises: a storage unit that stores evaluation data at the time of steering according to the electric power steering control device; steering situation data that specifies a situation under which the evaluation data is performed and evaluated, wherein the steering situation data is linked to the evaluation data; and control parameter data of the electric power steering control device.wherein the control parameter data are linked to the steering situation data and the evaluation data and cause the evaluation data to be expressed under the linked steering situation data, and adaptation procedure data of the control parameter data, wherein the adaptation procedure data are data relating to the control parameter data and are used to delete an evaluation content of the linked evaluation data; a first input unit that inputs the evaluation data; a second input unit that inputs the steering situation data; a read unit that reads the adaptation procedure data of the control parameter data relating to the input data from the storage unit based on the evaluation data and the steering situation data input by both input units; and an output unit that outputs the read adaptation procedure data of the control parameter data. List of patent literature

[0003] Patent Literature 1: JP 2002-173042 A Summary of the invention: Technical problem

[0004] The invention described in patent document 1 offers room for improvement in the selection of control characteristics. Solution to the problem

[0005] According to the first aspect of the present invention, a control characteristic selection device for searching for a control characteristic of an actuator that contributes to a driving characteristic of a moving body comprises a database in which a plurality of combinations of the control characteristic and a sensory evaluation rating are stored, a sensory evaluation rating unit for inputting the control characteristic as a current characteristic and outputting the sensory evaluation rating as a current evaluation rating, an input unit for receiving a target value that is the sensory evaluation rating to be targeted, and a selection unit configured to select the control characteristic from the control characteristics stored in the database.Select a characteristic where the sensory evaluation score is similar to the target value and similar to the current characteristic.

[0006] According to the second aspect of the present invention, a control characteristic selection method, executed by a computer capable of searching a database containing a plurality of combinations of a control characteristic of an actuator contributing to a driving characteristic of a moving body and a sensory evaluation rating of the control characteristic, comprises a sensory evaluation rating process for inputting the control characteristic as a current characteristic and outputting the sensory evaluation rating as a current evaluation rating, an input process for receiving a target value that is the sensory evaluation rating to be targeted, and a selection process for selecting from the control characteristics of the control characteristic stored in the database.in which the sensory evaluation score is similar to the target value and is similar to the current characteristic, as a selected characteristic, wherein the selection process selects as the selected characteristic the control characteristic that meets the target value and has a small difference in the characteristic value from the current characteristic. Advantageous effects of the invention

[0007] According to the present invention, it is possible to select a control characteristic that does not easily cause a sudden change in driving comfort while fulfilling the requirements of a user. Brief description of the drawings [ Fig. 1] Fig. Figure 1 is a configuration diagram of a control characteristic selection device. [ Fig. 2] Fig. Figure 2 is a diagram that represents a hardware configuration of a control characteristic selection device. [ Fig. 3] Fig. Figure 3 is a diagram that illustrates an example of a sensory evaluation assessment. [ Fig. 4] Fig. Figure 4 is a diagram that shows an example of data stored in a database. [ Fig. 5] Fig. Figure 5 is a diagram that represents a relationship between a control characteristic and a parameter. [ Fig. 6] Fig. Figure 6 is a diagram describing the learning of a neural network that forms a control characteristic selection unit. [ Fig. 7] Fig. Figure 7 is a diagram that represents a configuration of a neural network in a case where the output layer contains a large number of nodes. [ Fig. 8] Fig. Figure 8 is a function configuration diagram of the control characteristic selection unit. [ Fig. 9] Fig. Figure 9 is a flowchart that represents a process of the control characteristic selection device. [ Fig. 10] Fig. Figure 10 is a diagram that shows a display example of the display unit. [ Fig. 11] Fig. Figure 11 is a diagram that represents an indication in a case where a tax characteristic is changed after the delivery of a vehicle in a first modification. [ Fig. 12] Fig. Figure 12 is a diagram that represents a restriction on the input of the target value. [ Fig. 13] Fig. Figure 13 is a configuration diagram of a control characteristic selection system according to the second embodiment. Description of the embodiments-First embodiment-

[0008] A first embodiment of a control characteristic selection device is described below with reference to the Fig. 1 to Fig. 10 described.

[0009] Fig. Figure 1 is a configuration diagram of the control characteristic selection device 1. In the present embodiment, a person using the control characteristic selection device 1 is referred to as a "user." In this embodiment, the user is assumed to be a person involved in the manufacturing process of the vehicle on which the control characteristic selection device 1 is mounted. The control characteristic selection device 1 comprises a sensory evaluation unit 101, a setting unit 103, a control characteristic selection unit 104, a database 102, an input unit 105, and a display unit 106. The current characteristic 51 is entered into the sensory evaluation unit 101 and the setting unit 103.The control characteristic is a characteristic of an actuator, for example, a steering system, suspension, or brake, that contributes to the driving characteristics of the moving body. The control characteristic can also be described as the magnitude of the second physical quantity that corresponds to the first. There is only one second physical quantity that corresponds to the first. The control characteristic can be expressed by a polynomial or similar expression, and a simplified representation of the control characteristic is called a "parameter," which will be described later.

[0010] The sensory evaluation unit 101 estimates the sensory evaluation score according to the input control characteristic and outputs the current evaluation score 52. The sensory evaluation unit 101 incorporates a neural network, as described later. The learning process, i.e., the weight determination, is performed in advance within the neural network. The database 102 stores a large number of combinations of control characteristics and sensory evaluation scores. The display unit 106 is a display device, such as a liquid crystal display, used to present information to a user. The display unit 106 shows the current evaluation score 52 output by the sensory evaluation unit 101 and the selected characteristic 54 output by the setting unit 103.The display unit 106 can also display the current characteristic 51 and the target value 53.

[0011] The input unit 105 is, for example, a keyboard or a mouse. The user enters the target value 53, which is a sensory evaluation score, using the input unit 105 while viewing the display unit 106. Because sensory evaluation scores are difficult to assess on an absolute scale, the user enters the target value 53 based on the current evaluation score 52 of the current characteristic 51. The setting unit 103 receives the current characteristic 51, the evaluation score 502 output by the sensory evaluation unit 101, and the target value 53 entered by the input unit 105. The input unit 105 and the display unit 106 can be implemented as a touch panel.

[0012] Fig. Figure 2 is a diagram illustrating a hardware configuration of the control characteristic selection device 1. The control characteristic selection device 1 includes a CPU 41 as a central processing unit, a ROM 42 as a read-only memory device, a RAM 43 as a read / write memory device, an input / output device 44 as a user interface, and a communication device 45. The CPU 41 loads a program stored in the ROM 42 into the RAM 43 and executes the program to perform the various processes described above.

[0013] The control characteristic selection device 1 can be implemented by a field-programmable gate array (FPGA), which is a rewritable logic circuit, or by an application-specific integrated circuit (ASIC), which is an application-specific integrated circuit, instead of the combination of CPU 41, ROM 42, and RAM 43. Alternatively, the control characteristic selection device 1 can be implemented by a combination of different configurations, for example, a combination of CPU 41, ROM 42, and RAM 43 and an FPGA.

[0014] The input / output device 44 is an interface for data input / output with a user. The input / output device 44 could be, for example, a display, a mouse, a keyboard, a touch panel, a speaker, a microphone, or the like. The communication device 45 enables communication with an external device containing the database 102. The communication device 45 can use any suitable communication standard.

[0015] The control characteristic selection device 1 can receive input from an external device via the communication device 45. For example, a touch panel or button can be mounted on the dashboard of the vehicle on which the control characteristic selection device 1 is mounted, and the control characteristic selection device 1 can receive an input action from the user performed on the dashboard. Additionally, the communication device 45 can communicate with a mobile device owned by the user, and the control characteristic selection device 1 can receive an input action from the user at the communication device 45. Furthermore, instead of providing its own input / output device 44, the device 1 can communicate with an external device and use this external device instead of the input / output device 44 via the communication device 45.And the external device can be used instead of the input / output device 44. In this case, the data output by the control characteristic selection device 1 will be displayed on the external device.

[0016] Fig. Figure 3 is a diagram that illustrates an example of a sensory evaluation assessment. Fig. 3(a) shows a case in which there are five evaluation elements, and Fig. 3(b) shows a case in which there is an evaluation element. In Fig. 3(a) A pentagonal radar map is used to display rating values ​​of five rating elements simultaneously. Fig. 3(b) A bar chart is used. The user can refer to the current evaluation score of 52 on the [page number] in Fig. Referring to the display shown in 2, he enters the target value 53 using the input unit 105 while viewing the display.

[0017] In this example, the sensory evaluation score ranges from 0 to 7 points, and a guideline is provided every 0.25 points. The numerical range of the sensory evaluation score is an example and may vary. In the example of Fig. 3(a) The range from 3.50 to 4.50 points with a high response frequency is enlarged, but the display procedure is not restricted. Additionally, a case where the number of rating items is one and a case where the number of rating items is five are presented, but there is no upper limit as long as the number of rating items is one or more.

[0018] Fig. Figure 4 is a diagram that illustrates an example of data stored in database 102. Database 102 stores a characteristic and an evaluation score 502 for each evaluation element in conjunction with each other. Fig. Figure 4, however, shows a case where there are multiple evaluation elements. If there is one evaluation element, as in Fig. As shown in 3(b), there is a value of the rating element that corresponds to each characteristic.

[0019] Fig. Figure 5 is a diagram that illustrates the relationship between a control characteristic and a parameter. The three curves 504-1 to 504-3, which are shown in Fig. Figure 5 shows different characteristics. These characteristics indicate the relationship between the steering angle (or the pinion angle during steering) and the frictional torque, with the x-axis representing the steering angle and the y-axis representing the frictional torque. All characteristics exhibit the same tendency: the frictional torque increases as the steering angle increases. In the present embodiment, the "steering characteristic" indicated by the curve is converted into the "parameter" by sampling the "steering characteristic" within a predetermined interval of values ​​along the horizontal axis.When the steering angle and friction torque control characteristics are converted into calculation parameters, as shown in this figure, the value of the friction torque is sampled in increments of a predetermined value of the steering angle, for example in increments of 1 degree, and the value of the friction torque at 0 to 90 degrees is set as the “parameter”.

[0020] Fig. Figure 6 is a diagram illustrating the learning process of neural network 1041, which forms the control characteristic selection unit 104. First, a test driver drives a vehicle to which pre-generated characteristics are applied and determines a sensory value. By repeating this process multiple times, a combination of a characteristic and a sensory value, i.e., training data for neural network 1041, is generated. Neural network 1041 includes an input layer 305, a hidden layer 306, and an output layer 307. The number of nodes in input layer 305 is uniquely determined by a parameter determination procedure. For example, in the diagram with reference to Fig. In example 5, the input layer has 305 91 nodes, since the number of nodes is one degree from 0 degrees to 90 degrees. Fig. Figure 6 shows only one layer of the hidden layer 306 for the sake of simplicity, but multiple hidden layers 306 may exist. The number of nodes in the hidden layer 306 is arbitrary. The number of nodes in the output layer 307 can be one, as shown in the figure, or it can be multiple, as described later.

[0021] During the learning phase of neural network 1041, parameters generated based on each characteristic are inputted into input layer 305, and the weight of neural network 1041 is determined such that the value of output layer 307 matches the sensory evaluation score determined by the test driver. A known method, such as an error backpropagation method, can be used to determine the weight. The configuration of hidden layer 306 can be determined through learning. Since the learning process is computationally intensive, it is desirable to use a device, such as a server, with high computational capacity and a relaxed power consumption constraint instead of the control characteristic selection device 1.

[0022] Fig. Figure 7 is a diagram representing a configuration of neural network 1041 when output layer 307 contains a large number of nodes. In this case, output layer 307 uses one-hot coding. That is, each node is pre-assigned a sensory evaluation value, only one of the nodes is set to "1", and all other nodes are set to "0". For example, if the number of nodes in output layer 307 is five and the evaluation score 502 is "3.50" to "4.50", the lower part of Fig. The combination shown in section 5 is obtained. For example, if the test driver evaluates the sensory value of a particular test characteristic as "3.50", the weight of neural network 1041 is learned such that only "Y1" at the top of output layer 307 is "41" and the others are "0".

[0023] It is important to note that it is not essential to use the actual vehicle and test driver to generate the training data. A simulator that simulates the operation of an actual vehicle can be used instead of the actual vehicle itself, and a sensor output can be used instead of a test driver determining a sensor value. That is to say, in addition to the combination of the actual vehicle and test driver, any combination of simulator and test driver, actual vehicle and sensor, and simulator and virtual sensor can be used.

[0024] In the case of a simulator and a test driver combination, the test driver operates the simulator and determines the sensor reading. In the case of a combination of an actual vehicle and a sensor, the output of the sensor mounted on the actual vehicle is classified according to predetermined categories to obtain a sensor reading. In the case of a simulator and a virtual sensor combination, a virtual sensor is provided in the simulator, simulating the vehicle's operation, and the sensor's output is classified using a predetermined classification to obtain the sensor reading.

[0025] Fig. Figure 8 is a functional configuration diagram of the control characteristic selection unit 104. The control characteristic selection unit 104 includes a presentation count setting unit 701, a mode setting unit 702, a control unit 703, and a readout unit 704. The presentation count command 58 is entered into the presentation count setting unit 701. The presentation count command 58 can be entered by the user from the input unit 105, or a predetermined value can be entered. The presentation count setting unit 701 outputs the presentation count 71 to the control unit 703 based on the presentation count command 58. For example, the presentation count command 58 is a natural number, and the presentation count setting unit 701 can output the presentation count command 58 to the control unit 703 as the presentation count 71, exactly as it is.

[0026] Mode command 59 is entered into mode setting unit 702. Mode command 59 can be entered by the user from input unit 105, or a predefined value can be entered. Based on mode command 59, mode setting unit 702 selects one of three modes as decision mode 72: a parameter similarity priority mode, a compromise influence minimum mode, and a label element priority mode. Mode command 59 is a value that specifies one of the three modes.

[0027] The presentation count 71, output by the presentation count setting unit 701, and the decision mode 72, output by the mode setting unit 702, set the actuation of the readout unit 704. The presentation count 71 sets the number of characteristics output by the readout unit 704. For example, if the presentation count 71 is "1", the readout unit 704 outputs one selected characteristic 54 that best meets the condition. If the presentation count 71 is "5", the readout unit 704 outputs the five best selected characteristics 54 that meet the condition. The differences in the actuation of the readout unit 704 by the decision mode 72 are as follows.

[0028] In parameter similarity priority mode, the readout unit 704 selects a characteristic that meets the target value 53 and exhibits a high degree of similarity to the current characteristic 51. Similarity is the difference between characteristic values, and a smaller difference between characteristic values ​​indicates a higher degree of similarity. For example, if characteristic 504-1, which is in Fig. As shown in Figure 5, the current characteristic is 51, and characteristics 504-2 and 504-3 meet the target value of 53. Therefore, characteristic 504-2 is selected as the most suitable characteristic. This is because characteristic 504-3 exhibits a smaller difference in the friction torque value compared to characteristic 504-1 at each steering angle.

[0029] The compromise influence minimum mode and the designated rating element priority mode described below are based on the premise that there are a multitude of rating elements in sensory evaluation assessment, as in Fig. 3(a) shown. One or more of the multiple evaluation elements are representative elements. The representative element can be designated by the user each time or can be predefined in input unit 105.

[0030] In the minimum compromise influence mode, the readout unit 704 selects a characteristic that has a value in the representative element closer to the target value 53 than to the current characteristic 51. If there are multiple characteristics with the same value for the representative element, a characteristic that minimizes the total change in the sensory evaluation score is selected for other evaluation elements than the representative element. If there are multiple characteristics where the sums of changes in the sensory evaluation score are the same, a characteristic where the change from the current characteristic 51 is small is selected as the chosen characteristic 54.

[0031] In the designated evaluation element priority mode, the selection unit 704 selects a characteristic whose representative element has a value exceeding the target value of 53. In this mode, no other evaluation score 502 is considered besides the representative element. If there are multiple characteristics whose representative element has a value exceeding the target value of 53, the characteristic with the representative element's value closest to the target value of 53 is selected. If there are multiple characteristics where the representative element's value is closest to the target value of 53, a characteristic with a small change from the current characteristic 51 is selected as the characteristic 54.

[0032] The control unit 703 transmits the search condition, based on the outputs of the presentation count setting unit 701 and the mode setting unit 702, to the readout unit 704. The readout unit 704 searches the database 102 for a characteristic that matches the search condition and outputs the characteristic to the control unit 703. The control unit 703 transmits the characteristics output by the readout unit 704 to the setting unit 103. The readout unit 704 searches the database 102 according to the mode command 59, set by the mode setting unit 702, and outputs the selected characteristic 54 by the number of the presentation count command 58, set by the presentation count setting unit 701. The operation of the readout unit 704 in each mode is as described above.

[0033] Fig. Figure 9 is a flowchart illustrating the processing of the control characteristic selection device 1. In step S801, the current characteristic 51 is input to the sensory evaluation unit 101 and the setting unit 103. In the following step S802, the sensory evaluation unit 101 calculates the current evaluation score 52 and outputs it to the display unit 106 and the setting unit 103, and the process continues with step S803. In step S803, the display unit 106 shows the current evaluation score 52. In the subsequent step S804, the input unit 105 accepts an input of the target value 53 from the user, and the process continues with step S805.

[0034] In step S805, the control characteristic selection unit 104 calculates the selected characteristic 54. In step S806, the display unit 106 shows the selected characteristic 54, which was calculated in step S805. In a subsequent step S807, the setting unit 103 determines whether the user has entered a new target value 53, that is, whether the user has entered a new target value 53 from the input unit 105 or not. If it is determined that the user has entered a new value, the process returns to step S805. If it is determined that the user has not entered a new value, the process ends in Fig. 9 processes shown.

[0035] Fig. Figure 10 is a diagram illustrating a display example of display unit 106. This figure shows an example where a mobile device is used as display unit 106. Fig. Figure 10(a) shows the current characteristic 51 and the current evaluation score 52. The current characteristic 51 can be entered by the user, or the current setting can be read automatically. If the user enters the information, it can be entered on the screen using a touch pen or similar device, or it can be read from a CSV file or similar.

[0036] In Fig. Figure 10(b) shows the target value 53 and the selected characteristic 54. In the upper part of Fig. In 10(b), the current evaluation score of 52 is indicated by a solid line, and the user can enter the target value of 53 as a relative evaluation with respect to the current evaluation score of 52. A touch panel or voice input can be used for this input. The in Fig. The ad shown in section 10 is merely an example, and a different layout may be used, or all of them may be used. Fig. 10 pieces of information shown can be displayed on one screen.

[0037] According to the first embodiment described above, the following effects can be achieved. (1) Control characteristic selection device 1 searches for the control characteristics of the actuators that contribute to the driving characteristics of the moving body.The control characteristic selection device 1 includes a database 102 that stores a plurality of combinations of a control characteristic and a sensory evaluation rating, a sensory evaluation rating unit 101 that inputs a control characteristic as a current characteristic 51 and outputs a sensory evaluation rating as a current evaluation rating 52, an input unit 105 that receives a target value 53 which is a sensory evaluation rating, and a control characteristic selection unit 104 that selects a control characteristic with a sensory evaluation rating that is similar to that of the target value 53 and similar to that of the current characteristic 51 as a selected characteristic 54 from the control characteristics stored in the database 102.Therefore, not only is the characteristic similar to the user-specified target value 53 selected, but also the characteristic similar to the current characteristic, so that a sudden change in ride comfort is less likely to occur while still meeting the user's requirement. (2) In parameter similarity priority mode, the control characteristic selection unit 104 selects as the selected characteristic target value 54 a control characteristic with a sensory evaluation rating that is similar to that of the target value 53 and has a small difference from that of the current characteristic 51. (3) The sensory evaluation score includes a multitude of score elements, including a representative element. In the minimum trade-off influence mode, the control characteristic selection unit 104 selects a control characteristic 54 in which the representative element of the sensory evaluation score is similar to the target value 53 and the overall change in the sensory evaluation score, other than the representative element, is minimized. (4) In the designated evaluation element priority mode, the control characteristic selection unit 104 selects as the selected characteristic 54 a control characteristic in which the representative element of the sensory evaluation score exceeds the target value 53, and among these characteristics the one whose representative element value is closest to the target value 53. (5) The control characteristic specifies a second physical quantity corresponding to the first physical quantity, for example, a frictional torque corresponding to the steering angle. The sensor evaluation unit 101 specifies parameters, as with reference to Fig. 5 described, for the evaluation of the current characteristic 51. That is, the value of the second physical quantity, when the first physical quantity is increased by a predetermined unit of value, is specified as a parameter, and the parameter is entered into the pre-learned neural network 1041. (6) The actuator is one of a steering system, a suspension system and a brake system. (Modification 1)

[0038] In the embodiment described above, it is assumed that the "user" who operates the control characteristic selection device 1 is involved in the vehicle's manufacturing process. However, the control characteristic selection device 1 can also be operated by a person who uses the vehicle after its delivery.

[0039] Fig. Figure 11 is a diagram that displays an indication of when the tax characteristics are changed after the vehicle has been delivered. Fig. 11(a) depicts a scene in which the control characteristic selector 1 is operated using the instrument panel 1001 of the vehicle on which the control characteristic selector 1 is mounted. Fig. Figure 11(b) shows a scene in which the control characteristic selection device 1 is operated using a mobile device 1004 owned by the user. Fig. 11(a) The communication device 45 and the control device of the instrument panel 1001 communicate with each other, so that the instrument panel 1001 functions as an input / output device 44. That is, the target value 53 and the selected characteristic 54 are displayed on the instrument panel 1001. In Fig. Figure 11(b) shows the target value 53 and the selected characteristic 54 in the mobile device 1004. The user can obtain the selected characteristic 54 by entering the target value 53 using the dashboard 1001 and the mobile device 1004.

[0040] Fig. Figure 12 is a diagram illustrating the limitation of the input range for the target value 53. If a user—not the vehicle manufacturer, but a general user—sets the target value 53 after delivery of the vehicle, it is desirable to narrow the range within which the target value 53 can be set for safety reasons. Modifying the characteristics of the suspension, steering, and similar components alters the driving feel and comfort. Furthermore, applying inappropriate changes can lead to unexpected vehicle behavior. Therefore, an adjustable range for the target value 53 is predetermined, and the user can enter the target value 53 within this adjustable range.

[0041] In the Fig. In the example shown in 12(a), the non-adjustable area is indicated by hatching. In the example shown in Fig. In the example shown in 12(b), adjustable values ​​are indicated by white circles. Fig. 12(a) and Fig. 12(b) the values ​​that can be set as the target value 53 are the same in both examples. (Modification 2)

[0042] In the embodiment described above, the control characteristic selection device 1 is configured by a hardware device as shown in Fig. Figure 2 shows that the functions of the control characteristic selection device 1 can be implemented by a variety of hardware devices. In this case, a group of devices with functions equivalent to those of the control characteristic selection device 1 can be referred to as a "control characteristic selection system". (Modification 3)

[0043] In the embodiment described above, as referred to in Fig. As described in section 5, the control characteristic is converted into a parameter by sampling the control characteristic in increments of a predetermined value on the horizontal axis, and the parameter is input into the neural network. The control characteristics, as described in Fig. Figure 5, however, can be treated as image data, and this image data can be fed into neural network 1041. The image is generated by defining predetermined minimum and maximum values ​​for both the friction torque and the steering angle. When this modification is applied, it is necessary to re-perform the learning process of neural network 1041. -Second embodiment-

[0044] A second embodiment of a control characteristic selection system is described with reference to Fig. 13. In the following description, the same components as those of the first embodiment are designated with the same reference numerals, and differences are mainly described. Points not specifically described are the same as those in the first embodiment. The present embodiment differs from the first embodiment mainly in that data is added to the database.

[0045] Fig. Figure 13 is a configuration diagram of a control characteristic selection system S1 according to the second embodiment. The control characteristic selection system S1 further includes a characteristic addition device 1200 in addition to the control characteristic selection device 1 according to the first exemplary embodiment. The characteristic addition device 1200 comprises a characteristic data generation unit 1201, a database management unit 1203, and a second sensory evaluation evaluation unit 1202. The actuation of the second sensory evaluation evaluation unit 1202 is similar to that of the sensory evaluation evaluation unit 101. Like the sensory evaluation evaluation unit 101, the second sensory evaluation evaluation unit 1202 includes a learned neural network.

[0046] The characteristic data generation unit 1201 generates a large number of control features, automatically generated control characteristics 1251. Automatically generated control characteristics 1251 can be generated randomly, or a general tendency, for example, a monotonically increasing tendency, can be specified in advance. The second sensory evaluation evaluation unit 1202 generates parameters from each of the automatically generated control characteristics 1251, inputs the parameters into the built-in neural network, and outputs automatically generated evaluation values ​​1252.

[0047] Automatically generated control characteristics 1251 are entered into database management unit 1203 by characteristic data generation unit 1201, and automatically generated evaluation values ​​1252 are entered by the second sensory evaluation unit 1202. Database management unit 1203 evaluates the validity of each automatically generated control characteristic 1251. If it is determined that there is no validity issue, database management unit 1203 adds a combination of automatically generated control characteristics 1251 and automatically generated evaluation values ​​1252 to database 102.The evaluation of the validity of the automatically generated tax characteristics 1251 by the database management unit 1203 includes a monotonically increasing tendency, an upper limit and a lower limit of a minimum value, an upper limit and a lower limit of a maximum value, an upper limit of a change amount, and the like.

[0048] The hardware configuration of the characteristic addition device 1200 can be the same as the hardware configuration of the control characteristic selection device 1, which is described in Fig. 2 is shown. However, since the characteristic addition device 1200 does not require any input from the user, the characteristic addition device 1200 cannot include the input / output device 44.

[0049] According to the second embodiment described above, the following effects can be achieved. (7) The database management unit 1203 adds to the database 102 a combination of an automatically generated control characteristic 1251, which is an automatically generated control characteristic, and an automatically generated evaluation value 1252, which is a sensory evaluation rating obtained by evaluating the automatically generated control characteristics 1251. Therefore, the number of records in the database 102 can be increased, and a new control characteristic, which is closer to the target value 53 than before, can be obtained. (Modification of the second embodiment)

[0050] If automatically generated evaluation values ​​1252 for automatically generated control characteristics 1251, which are added to database 102, do not match the sensory evaluation value captured by the user, the user can perform the following process. That is, the user adds the automatically generated control characteristics 1251 and the sensory evaluation value captured by the user as a new combination to the training data of sensory evaluation unit 101 and the second sensory evaluation unit 1202. This improves the estimation accuracy of sensory evaluation unit 101 and the second sensory evaluation unit 1202.

[0051] In each of the embodiments and modifications described above, the configuration of the functional block is merely an example. Some functional configurations, represented as separate functional blocks, may be integrally configured, or a configuration represented by a functional block diagram may be subdivided into two or more functions. Furthermore, some of the functions of each functional block may be contained within another functional block.

[0052] In the embodiments and modifications described above, the program is stored in ROM 42. However, the program can also be stored in memory unit 46. The control characteristic selector 1 can include an input / output interface (not shown), and a program can be read from another device via the input / output interface and a medium that can be used by the control characteristic selector 1 if necessary. Here, the medium refers, for example, to a storage medium that can be attached to and detached from the input / output interface, or a communication medium, i.e., a wired, wireless, or optical network, or a carrier wave or digital signal propagating through the network. Some or all of the functions implemented by the program can be implemented by a hardware circuit or an FPGA.

[0053] The embodiments and modifications described above can be combined with one another. Although various embodiments and modifications have been described above, the present invention is not limited to these. Other embodiments conceivable within the scope of the technical concept of the present invention are also included within its scope. Reference symbol list 1 characteristic selection device 44 Input / output device: 45 Communication device: 51 Current characteristics: 52 Current evaluation rating: 53 Target value: 54 Selected characteristics: 58 Presentation count command: 59 Mode command: 101 Sensory Evaluation Assessment Unit: 102 Database: 103 Setting unit: 104 Control characteristic selection unit: 105 Input unit: 106 Display unit: 502 Evaluation assessment: 703 Control unit: 704 Reading unit: 1200 Characteristic Addition Device: 1201 Characteristic Data Generation Unit: 1203 Database Management Unit: QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2002-173042 A

[0003]

Claims

[1] Control characteristic selection device for searching for a control characteristic of an actuator that contributes to a driving characteristic of a moving body, comprising: a database in which a multitude of combinations of control characteristics and a sensory evaluation rating are stored; a sensory evaluation unit for sensory evaluation assessment for inputting the control characteristic as a current characteristic and outputting the sensory evaluation assessment as a current evaluation assessment; an input unit for receiving a target value, which is the sensory evaluation assessment for aiming; and a selection unit that is configured to select, from the control characteristics stored in the database, the control characteristic in which the sensory evaluation rating is similar to the target value and which is similar to the current characteristic, as a selected characteristic. [2] Control characteristic selection system according to claim 1, wherein the selection unit selects the control characteristic in which the sensory evaluation assessment is similar to the target value and the difference between the control characteristic and the current characteristic is small. [3] Control characteristic selection system according to claim 1, wherein the sensory evaluation rating includes a plurality of rating elements including a representative element, and wherein the selection unit selects the control characteristic in which the representative element of the sensory evaluation rating is similar to the target value and the totality of changes in the sensory evaluation rating that are not the representative element is minimized. [4] Control characteristic selection system according to claim 1, wherein the sensory evaluation rating includes a plurality of rating elements including a representative element, and wherein the selection unit selects the control characteristic in which the representative element of the sensory evaluation rating exceeds the target value and a value of the sensory evaluation rating of the representative element is closest to the target value. [5] Control characteristic selection system according to claim 1, wherein the control characteristic specifies a second physical quantity corresponding to a first physical quantity, and wherein the sensory evaluation evaluation unit specifies as a parameter a value of the second physical quantity when the first physical quantity is increased by a predetermined unit of value to evaluate the current characteristic, and uses the parameter as an input to a previously learned neural network. [6] Control characteristic selection system according to claim 1, wherein the actuator is one of a steering, a suspension and a brake. [7] Control characteristic selection system according to claim 1, further comprising a characteristic addition device which adds to the database a combination of an automatically generated control characteristic, which is the automatically generated control characteristic, and an automatically generated evaluation value, which is the sensory evaluation rating obtained by evaluating the automatically generated control characteristics. [8] Control characteristic selection procedure performed by a computer capable of searching a database containing a multitude of combinations of a control characteristic of an actuator contributing to a driving characteristic of a moving body and a sensory evaluation of the control characteristic, comprising: a sensory evaluation assessment process for inputting the control characteristic as a current characteristic and outputting the sensory evaluation assessment as a current evaluation assessment; an input process for receiving a target value, which is the sensory evaluation score to be targeted; and a selection process for choosing from the control characteristics stored in the database, in which the sensory evaluation rating is similar to the target value and which is similar to the current characteristic, as a selected characteristic, where the selection process selects the control characteristic that meets the target value and has a small difference in the characteristic value from the current characteristic.

Images