WO2014141351A1 - 自律制御装置 - Google Patents
自律制御装置 Download PDFInfo
- Publication number
- WO2014141351A1 WO2014141351A1 PCT/JP2013/056559 JP2013056559W WO2014141351A1 WO 2014141351 A1 WO2014141351 A1 WO 2014141351A1 JP 2013056559 W JP2013056559 W JP 2013056559W WO 2014141351 A1 WO2014141351 A1 WO 2014141351A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- constraint
- control logic
- autonomous
- control device
- expected
- Prior art date
Links
- 238000005259 measurement Methods 0.000 claims abstract description 20
- 238000006243 chemical reaction Methods 0.000 claims description 38
- 230000002401 inhibitory effect Effects 0.000 claims 1
- 230000005764 inhibitory process Effects 0.000 claims 1
- 230000002123 temporal effect Effects 0.000 claims 1
- 230000004044 response Effects 0.000 abstract description 15
- 238000000034 method Methods 0.000 description 60
- 230000008569 process Effects 0.000 description 48
- 238000012545 processing Methods 0.000 description 30
- 230000007704 transition Effects 0.000 description 20
- 238000004891 communication Methods 0.000 description 13
- 238000013461 design Methods 0.000 description 6
- 101000795744 Homo sapiens TPA-induced transmembrane protein Proteins 0.000 description 5
- 102100031626 TPA-induced transmembrane protein Human genes 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 238000013459 approach Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
- G05B13/026—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system using a predictor
-
- 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
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/196—Recognition using electronic means using sequential comparisons of the image signals with a plurality of references
- G06V30/1983—Syntactic or structural pattern recognition, e.g. symbolic string recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
Definitions
- the present invention relates to an autonomous system operating in an open environment and an autonomous operation control device that determines the operation of the autonomous system.
- an autonomous mobile robot heading for a destination while satisfying the condition of a sound operation to avoid a collision with a person moving around indefinitely as shown in Patent Document 1 or
- an autonomous work system that executes a predetermined work
- a totally autonomous traveling type vehicle that uses a route generation function to a remote destination in combination.
- Patent Document 2 a robot having an autonomous movement function in a closed environment is presented.
- Patent Document 3 includes an autonomous work system in which a user steers remotely via a communication path.
- the remote pilot can be configured to be able to select between the remote control function and the autonomous operation function in real time, and the remote pilot can pre-program the expected operation or instruct the operation on a low frequency as needed.
- An example is described where the work system implements the indicated expected behavior.
- Patent Document 4 as a semi-autonomous system in which the pilot rides the system and directs an expected operation directly, an intervention control is performed to detect a distance between vehicles and avoid a collision between adjacent vehicles.
- an autonomous system specialized for driving assistance which has a function.
- the operation mode and the corresponding control logic are incorporated in the form of a program etc., and it is difficult for a third party to directly refer to or evaluate at the time of operation.
- the information obtained through observation is ultimately limited to the behavior realized through the control logic.
- the operation mode in which the soundness of the operation is not verified is that the control logic designed corresponding to the operation mode assumed in the design stage satisfies the soundness of the operation. In other words, it corresponds to the operation in the state that does not satisfy the premise assumed implicitly.
- Such an operating condition is not assumed at the time of design, and thus will be corrected one by one if it is discovered through real machine testing.
- the coverage of the operating environment actually encountered and the soundness of the corresponding control logic are considered. Hard to guarantee.
- there is a problem that the number of operation modes to be assumed and the number of control logics to be designed are combined and explosively increased according to the variety of external factors accompanied by setting or changing of the non-intrusion area.
- an operation control logic for realizing an operation adapted to an internal factor of the own system and an external factor in the open external environment acquired by the external world measurement means based on the operation control logic stored in the storage device.
- the autonomous control device that generates and instructs the drive device to operate in the open external environment includes the operation control logic and the restriction condition corresponding to the external factor and the internal state of the own system, which the autonomous control device has acquired from the external world measurement means.
- the operation control logic and the control logic are generated from the constraints related to the expected operation and the soundness of the operation required for the own system each time the system is sequentially rebuilt, and the generated operation control logic and the external world measuring means
- An autonomous operation control device that writes information acquired from the device to the recording device and instructs the operation of the own system based on the control logic, and the autonomous operation control Based on the determination result of the soundness of the expected operation and operation by the device, comprising the expected operation determining unit for instructing the expected behavior that is required of the local system.
- a sound operation is realized by sequentially generating control logic whose operation soundness is verified by an operation mode adapted to a changing external factor, and at the same time, the generated control logic is recorded and presented to operate.
- a third party can confirm the soundness of the operation by referring to the decision process.
- Configuration of autonomous system Processing flow of autonomous operation control device Definition of closed area Representation of constraints and formulas Control logic L definition Representation of motion by constraint form Representation of behavioral rules and expected behavior by relaxed conditional constraint form Processing flow of construction of data structure indicating connection relationship between partial areas 2 Classification and judgment method of connection relation of closed area Example of division into partially closed regions Functional block diagram of constraint form conversion means A Conversion rules used by the constraint form conversion means A Processing flow of constraint format conversion means A Functional block diagram of constraint format conversion means B Example of a program that indicates expected behavior Control structure of program that indicates expected behavior decision logic Processing flow of constraint format conversion means B Processing flow of expected behavior determination logic Generation of operation control logic T and control logic L Analysis of connection between closed regions List of action control logic that can be executed Movement operation between two closed areas instructed by the operation control logic exemplified in FIG.
- Control of autonomous system by remote pilot Internal configuration diagram of autonomous traveling system Autonomous traveling system that operates to enter an intersection Contents of the autonomous traveling control program used in the third embodiment Conversion rules used by the constraint form conversion means A in the third embodiment Relaxed conditional constraint form K output by the constraint form conversion means A in the third embodiment Operation control logic generated when the signal value is not red in the third embodiment
- FIG. 1 shows a block diagram of an autonomous system.
- the autonomous system 01 operating in an open external environment includes an autonomous control device 010, an external measurement unit 011, an internal state measurement unit 012, a presentation device 013, and a drive device 014.
- the autonomous control device 010 includes an autonomous operation control device 0101, a recording device 0102, an expected operation determination logic 0103, a constraint type conversion unit A (0104), and a constraint type conversion unit B (0105).
- Constraint type conversion means A (0104) converts information of the open external environment obtained through the external world measurement means 011 into a restriction form and passes it to the autonomous operation control device 0101.
- the recording device 0102 stores the generated record 0102 of the operation control logic generated by the autonomous operation control device 0101.
- the autonomous operation control device 0101 presents the generated record 0102 of the operation control logic to the presentation device 013 as necessary.
- the expected operation determination logic 0103 operates based on the expected operation instruction program 01031, and the constraint format conversion means B 0105 converts the expected operation determination logic 0103 into a constraint format and passes it to the autonomous operation control device 0101.
- the autonomous operation control device 0101 determines the satisfiability of the expected behavior converted to the constraint format, and returns the result as satisfiable, satisfiable, or processing complete.
- the autonomous operation control device 0101 outputs a control command to the drive device 014 after the process shown in FIG.
- FIG. 2 shows a process flow performed by the autonomous control device 010.
- step 021 a list of constraint types output from the constraint type conversion means A (0104) and the constraint type conversion means B (0105) is received and a constraint type to be processed is set.
- the list of constraint types received in step 021 is stored in the recording device 0102.
- step 023 operation control logic T of the entire autonomous system 01 that satisfies the list of constraint types set in step 021, and control logic L for calculating control commands to be output to the drive device 014 are generated. Then, in step 024, it is verified that the operation control logic T and the control logic L which satisfy all the constraint types set in step 021 can be realized. If it is determined that the verification result is feasible, the process proceeds to step 026. In step 026, the operation control logic T and the control logic L are stored in the recording device 0102. On the other hand, if it is determined that the method can not be realized, the process proceeds to step 025. The list of constraint types recorded in step 022 is read out, and at least a part of the constraint types set in step 021 is relaxed. For example, if the constraint that can not be satisfied in step 024 can be relaxed, it is replaced with a relaxed constraint type. Subsequently, the process proceeds to step 021 to repeat the process.
- step 026 After recording the operation control logic T and the control logic L in step 026, the process proceeds to step 027, and a control command is output to the drive device 014 based on the control logic L. Then, in step 028, it is checked whether the constraint type set in step 021 is not updated. If the constraint type is not updated, the process returns to step 027 to continue the operation based on the same control logic L, and when updated, the process returns to step 021.
- FIG. 3 shows the definition of a constraint that represents a closed region REGION (Y) by combining a plurality of constraints (COND_0 (Y) to COND_3 (Y)).
- This constraint is constructed as a conjunction of individual constraints which determine the inside of the half plane group which gives the face of the closed region in the parameter space Y representing the internal state.
- FIG. 4 shows operators and their notations for combining complex constraints by combining the constraints.
- a method of expressing the operation output by the control logic L will be described with reference to FIG. 5 using the constraints indicating the closed region and the conditional constraints.
- the point P0 in the closed region R0 & (! R1) passes through the point P1 in the closed region R0 & R1 and then the closed region!
- the action of moving to point P2 in R0 & R1 is represented as Action.
- FIG. 6 shows the definition of a constraint form (operator) having the constraint condition Ci as an argument.
- the operator X [C0] for the constraint C0 is a constraint form that expresses conditional behavior. This is a constraint form that expresses an operation that satisfies the constraint condition C0 serving as an argument (the evaluation value of the constraint condition C0 is TRUE) in the following order with respect to the appropriately set sequence.
- the operator [C0] R [C1] for the constraints C0 and C1 is a constraint form that represents conditional operations.
- the operator [C0] W [C1] for the constraints C0 and C1 is a form of constraint that expresses conditional operations.
- FIG. 7 shows the definition and expected behavior of the relaxed conditional constraint form having the constraint form Ci as an argument.
- Each relaxed conditional constraint type is given by the logical sum of one or more constraint types, and is classified based on the relaxability of the constraint on the ordered sequence. Then, the presentation content to the presentation device 013 corresponding to the case of applying the relaxation condition is defined for each relaxation condition-bound type.
- the autonomous operation control device 0101 receives, from the constraint type conversion means A (0104) and the constraint type conversion means B (0105), data of a constraint type represented by these operators and expression forms.
- binary coding is performed by setting the inside including the outer edge of the closed region to 1 and the outside to 0. Then, according to the processing flow shown in FIG. 8, in the situation where the closed region group X has already been defined in S901, when a single closed region Y is newly added in S902, the closed region group X is added as shown in FIG.
- the two closed regions of Case 1 coincide, the two closed regions of Case 3 are separated if the two closed regions of Case 2 partially overlap and are in contact with each other, based on the connection relationship between If the two closed regions of Case 4 are in the inclusion relationship, if the two closed regions of Case 5 are in the covering relationship, if one closed region of Case 6 divides the other closed region, It is classified into either.
- the constraint condition [C0 & C1] is set using the constraint condition C0 for expressing the closed region X and the constraint condition C1 for expressing the closed region Y.
- SAT a solution that satisfies
- UNSAT UNSAT
- the constraint condition [! There is no solution that satisfies C0 & C1]
- constraints [! C0 &! It corresponds to the case where there is a solution that satisfies C1].
- the criteria for satisfying solution shown in FIG. 9 may be applied to each combination of constraints.
- connection relationship of the partially closed regions divided by the closed regions is updated, and in S905, the region is added based on the update result.
- FIG. 10 shows an example of sequentially adding the constraint condition C1 indicating the closed region R1 and the constraint condition C2 indicating the closed region R2 to the constraint condition C0 indicating the closed region R0 in the process illustrated in FIG. .
- R1 is newly set as the closed region ID.
- S903 it is determined that the closed regions R0 and R1 are classified into the case 2 shown in FIG. 9, and the closed region R0 is divided into R0-0 and R0-1 in S904, and the closed region R1 is R1.
- R0-0 and R0-1 and R1-0 and R1-1 are originally continuous, these connections are added to update the connection relationship, and R0-1 and R1-1 are case 1 (2 Since the closed regions correspond to one another), the connection relationships are updated by adding the connection relationships of R0-1 to R1-0 and R1-1 to R0-1. The process is similarly performed when the closed region R2 is added.
- FIG. 11 shows functional blocks of the constraint format conversion means A (0104).
- the constraint type conversion means A (0104) As shown in FIG. 12, the classification of the recognition object, the list of constraints defined in relation to each object, the state value of the recognition object, and the recognition object A template of a relaxation conditional constraint type that indicates the corresponding operation rule is provided in advance as a conversion rule.
- a relaxation conditional constraint form is generated for each individual recognition object, and the logical product K (01045) is added to the autonomous operation control device 0101 together with the constraint form list. Output.
- the external recognition unit 01041 receives the information M (0111) of the recognition target acquired by the external measurement unit 011, and the external recognition unit 01401 classifies the recognition target according to the conversion rule template ( S1701). Then, a restriction condition list is set for each classification of the recognition target (S1702), and the state value for each classification of the recognition target is obtained and set (S1703), and the restriction generation unit 0104 is used to recognize the recognition target. In each case, a relaxation conditional constraint form is specified that indicates the operation rule specified by the template (S1704). Then, the logical product K of the set relaxation conditional constraints is generated and output to the autonomous operation control device 0101 (S1705).
- the state value of the recognition target object set through the external world recognition unit 01041 is combined with the internal state value of the autonomous system 01 acquired through the internal state measurement unit 012 to make the autonomous operation control device 0101 as an observation point of the current state. Output.
- the external world measurement means 011 is a means for acquiring information necessary when the autonomous system 01 acts in an open external environment, and is not limited to a passive device represented by an imaging device or a measure sensor.
- the input data to the external world recognition unit 01041 may be received from the system in the open external environment through some communication path.
- the expected operation determination logic 0103 outputs the relaxation conditional constraint type E to the autonomous operation control device 0101 to the autonomous operation control device 0101 through the constraint type conversion means B (0105), The process until the expected operation control logic 0103 resets the output value will be described based on the response (SAT: saturable, UNSAT: non-satisfied, COMP: completed).
- the expected operation determination logic 0103 internally includes an expected operation instruction program 01031 for generating an expected operation as shown in the example of description shown in FIG. 15, and the operation control shown in FIG. In accordance with the logic, a list of relaxation conditional constraints is instructed to the constraint format conversion means B. The expected operation determination logic 0103 continues the operation from the operation start to the operation stop condition according to the process flow shown in FIG. 17 with reference to the operation ID shown in FIG.
- the minimum unit of the program that instructs the expected operation is defined for each operation ID, and an operation described in the relaxation conditional constraint form 010311, an operation completion determination condition described as a constraint condition 010312,
- the transition destination designated for each of the three types of responses (SAT, UNSAT, COMP) returned from the autonomous operation control device 0101 to the expected operation determination logic 0103 The operation is performed based on the control logic composed of the operation IDs (in order of 010313, 010314, and 010315).
- the initial value of the action ID is described as IDS
- the final state is described as IDE.
- FIG. 17 shows the process flow of the constraint type conversion means B (0105).
- the expected operation E (010311) of the relaxed conditional constraint format corresponding to this operation ID, the judgment condition of operation completion 010312, and the branch after obtaining the response
- the transition destination operation ID (010314) after completion of the operation is acquired from the conditions (S2202, S2203, S2204).
- an operation 010311 of a relaxation conditional constraint format corresponding to the acquired transition destination operation ID after operation completion is acquired (S2205), and a relaxation conditional constraint format E instructing the expected operation and the constraint format E are configured.
- a list of constraint conditions and the operation completion condition 010312 are output to the autonomous operation control device 0101 (S2206, S2207, S2208).
- FIG. 18 shows the process of responding to the autonomous operation control device 0101 in the specified operation ID.
- the operation ID is set to the initial value IDS, and the processing flow of the constraint format conversion means B shown in FIG. 17 is executed (S2302), and a response from the autonomous operation control device is awaited (S2303).
- SAT satisfiable
- the process when satisfiable is executed, and then the process returns to S2303 to continue to wait for a response.
- the response value is complete (COMP)
- processing S2306 is executed.
- the response value is unsatisfiable (UNSAT)
- processing S2305 is performed when the response value is unsatisfiable.
- FIG. 19 shows a flow of processing for generating the operation control logic T and the control logic L. This corresponds to the process of step 023 in FIG.
- an initial process is performed to analyze the connection relationship between partially closed areas and closed areas divided by overlapping of the closed areas.
- the relaxed conditional constraint form K passed to the autonomous operation control device 0101 in FIG. 11 and the constraint condition list as an argument
- the relaxed conditional constraint form E passed to the autonomous operation control device 0101 in FIG. Get the constraint list that becomes
- step 02301 the correspondence between the partially closed area and the constraint condition and the connection relation between the partially closed areas are generated along the processing flow shown in FIG. 8 (step 02301), and the logical state value is generated.
- an operation rule A which is a relaxation conditional constraint form is newly set as a logical product of the relaxation conditional constraint forms K and E (step 02305).
- FIG. 21 is a list for recording what can be realized by the operation control logic T generated in step 023.
- the first is an empty set, and the RINIT is added to the list of operation control logic IDs.
- the list of motion control logics shown in FIG. 21 is not an empty set, the processing that follows is continued.
- step 02390 one operation control logic is selected by giving priority to those to which the "relaxation condition given to the expected operation" is not applied. As an example, one having a small value of the operation execution order ID required to complete the operation may be selected.
- Each operation control logic T_i is specified as a chain of closed area IDs to be transitioned for each operation execution order ID, and as long as the definition of closed areas shown in FIG. The truth value of the condition is to be inverted. Using this characteristic, in accordance with the procedure shown in FIG. 19, in step 02391, a control logic L corresponding to each transition between closed regions is determined as a constraint condition.
- the autonomous operation control device 0101 returns a response of satisfiable (SAT) to the expected operation control logic 0103 (step 02307), and ends the operation.
- SAT satisfiable
- step 02304: No If operation control logic that does not satisfy the operation completion determination condition (010312) remains in the list shown in FIG. 21 (step 02304: No), the process of generating the operation control logic T is continued.
- one operation control logic that does not satisfy the operation completion determination condition (010312) is selected from the list of FIG. 21 and is referred to as TTMP.
- the closed region ID corresponding to the maximum operation execution order ID shown in FIG. 21 is obtained for TTMP, the true / false value of each constraint condition constituting the closed region is evaluated, and binary coding is performed.
- the transition destination closed regions having a connection relationship from the region ID it is evaluated whether there is any one satisfying the operation rule A set in step 02305.
- step 02303 when the list of operation control logic T shown in FIG. 21 becomes an empty set, the autonomous operation control device 0101 returns a response of unsatisfiable (UNSAT) to the expected operation control logic 0103 (step 02308), end the operation.
- UNSAT unsatisfiable
- step 02306 the transition destination closed regions are updated to the operation control logic added to the operation control logic TTMP.
- the relaxation condition V is shown in FIG. It records in the corresponding part corresponding to operation control logic TTMP.
- step 02307 or 02308 is reached and terminated.
- the autonomous operation control device 0101 ends the processing step 023.
- FIG. 22 shows a problem of generating an operation in which the closed region ID becomes R2-2 within the range where the region ID does not become EXTERIOR from the state of the closed region ID being R0-0 shown in FIG.
- operations for R0-0 to R0-1, R1-0, R1-1, R1-2, R2-1, and finally R2-2 are generated, and the final operation is performed.
- the operation control logic shown in FIG. 21 is obtained.
- step 02307 is reached in FIG. 19, it is regarded as verification success in step 024, and the process proceeds to step 026.
- the list of the operation control logic T (02392) generated according to the processing flow shown in FIG. 19 and all the control logic L_i associated with each operation control logic T_i are stored in the recording device 0102. Then, in step 027, the control logic L is executed to output a control command to the drive unit 014.
- a dynamic model of the autonomous system 01 to be controlled is acquired. This may be obtained in advance by system identification or may be generated during operation.
- step 02310 the control logic L generated at step 02391 is obtained.
- step 02311 the true / false value of the constraint condition list shown in FIG. 12 is evaluated using the observed value 01044 in the current state, and the closed region ID to which the state value belongs is calculated.
- step 02312 of the control logic L_i acquired in step 02310, the control logic L [k] to be immediately executed from the current state is selected.
- the purpose of use of the presentation device is to present the unsatisfiable constraint form generated in the process of executing the expected operation to the outside of the autonomous system, and is not limited to the form represented by the visual display device.
- the presentation content may be transmitted to other systems existing in the open external environment through the communication path.
- steps 02313 and 02314 logical state values corresponding to the individual constraint conditions giving the transition source closed region and the transition destination closed region are acquired, and in step 02315, the constraint condition Cp whose true / false value is inverted is specified.
- This restriction condition defines a closed region which enters or exits in the process of moving from the transition source closed region to the transition destination closed region according to the control logic L [k].
- step 02316 The true / false value of the constraint condition described in step 02316 is evaluated, and the process transitions to step 02317 or step 02318.
- Step 02317 corresponds to the movement from the outside to the inside of the closed region Cp.
- the condition described in step 02317 that is, the expanded closed region Cp becomes expanded so as to include the observed value Y (t) of the current state for the first time
- Y (t) the observed value of the current state for the first time
- step 02319 it is determined whether or not there is a control input sequence such that the observed value 01044 in the current state approaches the inside of the closed region Cp. If it exists, the calculated control input value is output to the drive device 014, and in step 02322, the judgment condition 010312 of the operation completion which should be satisfied at the end of execution of the control logic L [k] is evaluated. If the operation completion condition is not satisfied, the process returns to step 02317. If the operation completion condition is satisfied, the process proceeds to step 02323, the operation completion (COMP) is transmitted to the expected operation determination logic 0103, and step 027 is ended. Conversely, if there is no control input sequence in step 02319, the control logic L can not be executed, and it corresponds to the operation control logic L selected in step 02310 from the list of operation control logics registered in FIG. Delete the operation control logic T_i.
- step 02325 operation control logic not including the control logic L [k] determined to be unexecutable in step 02319 is searched from FIG. If there is, the process goes to step 02327 to reset the operation control logic and the control logic, and returns to step 02311. If not, the process proceeds to step 02326, returns unsatisfiable (UNSAT) to the expected operation control logic 0103, and ends the processing.
- UNSAT unsatisfiable
- step 02318 corresponds to an operation of coming out from the inside of the closed region Cp.
- the condition described in step 02318, that is, the reduced closed region Cp is the maximum of the reduction parameter s including the observed value Y (t) in the current state.
- the integer value By calculating the integer value, an operation of leaving the transition source closed region while realizing the transition destination closed region is realized.
- step 02320 it is determined whether or not there is a control input sequence such that the observed value 01044 in the current state approaches the inside of the closed region Cp.
- the calculated control input value is output to the drive device 014, and in step 02328, the judgment condition 010312 of operation completion which should be satisfied at the end of execution of the control logic L [k] is evaluated.
- step 02318 If the operation completion condition is not satisfied, the process returns to step 02318. If the condition is satisfied, the process proceeds to step 02323, the operation completion (COMP) is transmitted to the expected operation determination logic 0103, and step 027 is ended.
- COMP operation completion
- control logic L can not be executed, and the operation corresponding to the control logic L selected in step 02310 from the list of operation control logics registered in FIG. Delete control logic T_i.
- step 02325 operation control logic not including the control logic L [k] determined to be unexecutable in step 02319 is searched from FIG. If there is, the process goes to step 02327 to reset the operation control logic and the control logic, and returns to step 02311. If not, the process proceeds to step 02326, and the expected operation control logic 0103 is returned with the response that it can not be satisfied (UNSAT), and the processing is terminated.
- FIG. 26 shows the operation record that the autonomous operation control device 0101 outputs to the presentation device 013 at step 02329 and the generation record 0102 of the operation control logic stored in the recording device 0102 in the whole process of FIG.
- At least the generation record 0102 of the operation control logic includes external measurement information M (0111), time-series data of the observed value 0104 in the current state, and the relaxation condition restriction form K (01045) indicating the operation rule shown in FIG. And a constraint condition list serving as an argument, a program 01031 instructing an expected operation, a relaxation conditional constraint form E (01032) instructing an expected operation, a constraint condition list serving as an argument, and the list shown in FIG.
- Operation 010311 corresponding to expected operation E determined to be unrealizable (UNSAT) in step 02308 or 02326, control logic L associated with each list of operation control logic T (02392) and each operation control logic Control logic L selected and executed at step 02310 in 027, step during operation Infeasibility determined to be deleted operation control logic T_i in 2324, and a control logic L_i associated with the operation control logic T_i.
- UNSAT unrealizable
- the healthy operation is defined.
- the operation mode is reconfigured during operation (step 02327) using a function (steps 023 and 024) that derives a combination that can satisfy both the required requirements and the expected operation, and the control logic for realizing the expected operation is concurrently processed.
- the autonomous operation control device 0101 to be generated the external factor dynamically changing in the open environment, the requirement for defining the soundness of the operation (01045 realized in the form of a relaxation conditional constraint), the expected operation (in a relaxed conditional constraint form) And a function to record the realized 0103) and the generated control logic (02392) on the recording device 0102 in a form that can be viewed by a third party, It has achieved an autonomous system 01 to operate while presenting to a third party in response.
- step 02308 If the autonomous operation control device 0101 inside the autonomous system 01 reaches the unsatisfiable (UNSAT) in step 02308 or step 02326, all the operations indicated by the expected operation determination logic 0103 become unsatisfiable as well.
- the operation ID inside the expected operation determination logic reaches [Stop (IDE)] and stops operation.
- the third party who evaluates the operation soundness of the autonomous system in particular the individual operations 010311 that the expected operation determination logic 0103 sequentially instructed are It is possible to know the process finally determined to be unrealizable (UNSAT) by the processing inside the autonomous operation control device 0101 (step 02324) and the factor that became unsatisfiable.
- UNSAT unrealizable
- the designer of the autonomous system 01 corrects the program 01031 instructing an expected operation to generate the constraint type E (01032) determined to be unsatisfiable, or the operation rule set in the constraint type setting unit 0104
- the problem is solved by means such as adding a new file in the setting unit 01043.
- FIG. 27 shows an aspect of an autonomous system having a function to be steered by the remote pilot 02 via a communication path.
- the autonomous system 01 configured as in the first embodiment is disposed at a remote place, and the operation instruction 0201 is issued as needed with reference to the external world measurement information and the internal state measurement information acquired through the communication path 0202, and the expected operation
- the decision logic is configured to convert to the expected operation (01032). If the operation control logic T can not be generated, the remote pilot substitutes the process executed in step 025 so that a sufficient solution for achieving the sound operation exists.
- Patent Document 3 shows a method of combining the case of operating in the fully autonomous mode and the case of operating by remote control in real time.
- the soundness of operation depends on factors such as the quality of the communication path, that is, the bandwidth and response delay time, and the blocking of the communication path.
- Such a configuration makes it possible to easily issue and update various expected operations without depending on the quality of the communication path, and ensure the operation soundness of an autonomous system operating at a remote location while remote control is performed.
- the availability at the time of operation can be greatly improved.
- the increased availability appears in particular in that the number of autonomous systems that can be manipulated by the remote pilot per person can be significantly increased beyond one.
- the autonomous system itself can execute functions related to the determination and solution of the operation soundness. Therefore, the remote pilot needs to perform remote control operation on the autonomous system only when the expected motion generated from the motion command is not satisfied. As a result, one remote pilot can transmit operation commands to a plurality of autonomous systems and be operated in parallel. (Example 3)
- Example 3 an example of a traveling control function of an autonomous traveling system in an open city environment to which the present invention is applied will be shown.
- FIG. 1 The block diagram of the autonomous traveling system 03 is shown in FIG.
- the autonomous traveling system 03 operating in the open external environment is composed of an autonomous traveling control device 030, an external measuring means 031, an internal state measuring means 032, a presentation device 033, and a driving device 034.
- the autonomous control device 030 is an autonomous traveling control The apparatus 0301, the recording apparatus 0302, the expected operation determination logic 0303, the restriction form conversion means A (0304), and the restriction form conversion means B (0305).
- FIG. 29 shows the appearance of the operating environment in the case where the autonomous traveling system 03 shown in FIG. 28 executes an operation for crossing the intersection with a signal toward the target point SGOAL.
- FIG. 30 shows the operation command described in the autonomous traveling control program 03031 in this case.
- the operation ID is ID 0
- an operation to go into the SGOAL is instructed from the initial state (030311 and 030312).
- the autonomous operation control device 0301 determines that the expected operation can be satisfied (030313)
- the expected operation is continued and the operation ID is maintained at ID0.
- the autonomous operation control device 0301 determines that the expected operation can not be satisfied (030315)
- the expected operation is cleared and the operation ID is set to IDE. Thereafter, from the operating environment shown in FIG. 29, the mode is shifted to a mode for generating an operation from the relaxed conditional constraint type K indicating the operation rule specified by the constraint type conversion unit A0304 through the external world measuring unit.
- FIG. 31 corresponds to the conversion rule used by the constraint type conversion means A shown in FIG. 12 in this embodiment. Take roads, traffic lights, intersections, pedestrians and vehicles as objects to be recognized.
- the output content to the display device 033 is set corresponding to the satisfiability of each relaxation conditional constraint type K.
- the autonomous motion control device (0301) in the autonomous traveling system in the operating environment shown in FIG. 29 corresponds to the processing flow (step 02301) in which the connection relationship between closed regions shown in FIG. 20 is analyzed.
- the connection relationship of the closed region set for each recognition object is analyzed by the processing described above.
- the external world recognition unit 03041 inside the restriction format conversion means A0304 for example, the road 0 in the straight direction, the road 1 in the cross direction, the signal 0 for traffic control, and As shown in FIG. 32, according to the format shown in FIG. 12, registration is performed as shown in FIG.
- the constraint condition generation unit 03042 sets a constraint condition group according to the contents defined in FIG. 31 for each recognition target.
- the restriction format setting unit 0304 sets the relaxation condition restriction format K in accordance with the contents defined in FIG. 31 for each recognition object.
- FIG. 26 shows operation control logic T corresponding to FIG. 21 in the operating environment shown in the present embodiment.
- FIG. 28 shows operation control logic for realizing the expected operation shown in FIG. 30 obtained through the process flow shown in FIG. 19.
- the control operation is started from the case where the initial state is inside the closed region R0-0-1, and the closed region R0-1-1 and the closed region R0-2-0-1, and finally the closed region R0-2 -1-1 represents the logic of an operation for transition to the inside.
- This operation satisfies the judgment condition of operation completion in the case where the operation ID is ID 0 shown in FIG. 30, that is, the condition that the constraint condition C3 becomes 1, and realizes the expected operation.
- control logic L [0] shown in FIG. 23 is used to generate the processing flow of control logic shown in FIG. 26, which corresponds to the process of transitioning from 0 to 1 in the operation order ID.
- the operation execution order ID shown in FIG. 30 corresponds to 3 as a process of entering the area with the intersection while staying in the area of road 0 while the operation is realized through The operation is continuously continued according to the processing flow shown in FIG. 23 until the determination condition (030312) of is satisfied. After that, as shown in FIG. 30, the operation ID is transited to IDE to stop the operation.
- the present invention can be used for a fully autonomous vehicle operating in an open environment including a public road, and an unmanned taxi and an unmanned car rental service to a passenger's waiting position. Further, it can be used for an automobile having an auto cruise function and a collision avoidance function which travel at a constant speed without contacting surrounding traveling vehicles. In addition, it is a working machine that controls through a communication path in a remote place typified by mine, deep underground, and seabed, and can be used for autonomously judging and operating so as to satisfy constraints on operational safety.
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Transportation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Business, Economics & Management (AREA)
- Mechanical Engineering (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Stored Programmes (AREA)
Abstract
Description
以下、本発明を用いた自律システムの第一実施例について説明する。図1に、自律システムの構成図を示す。開放外部環境で動作する自律システム01は、自律制御装置010、外部計測手段011、内部状態計測手段012、提示装置013、及び駆動装置014を備えている。自律制御装置010は、自律動作制御装置0101、記録装置0102、期待動作決定論理0103、制約形式変換手段A(0104)、及び制約形式変換手段B(0105)で構成される。
この時、閉領域Cpを適当な比率γで拡大することで、ステップ02317にて記載の条件、つまり拡大した閉領域Cpが、現状態の観測値Y(t)を初めて包含するようになる拡大パラメータsの最小整数値を算出することで、遷移元閉領域に留まりながら、遷移先閉領域に侵入する動作を実現する。
(実施例2)
図27は、通信経路を介して遠隔操縦者02が操縦する機能を有する自律システムの一態様を示す。
(実施例3)
以下、第3の実施形態として、本発明を適用した開放都市環境における自律走行システムの走行制御機能の例を示す。
010 自律制御装置
011 外界計測手段
012 内部状態計測手段
013 提示装置
014 駆動装置
0101 自律動作制御装置
0102 記録装置
01021 動作制御論理の生成記録
0103 期待動作決定論理
01031 期待動作指示プログラム
0104 制約形式変換手段A
01041 外界認識部
01042 制約形式生成部
01043 制約形式設定部
01044 現状態の観測値
01045 緩和条件付き制約形式K
0105 制約形式変換手段B
02 遠隔操縦者
0201 動作命令
0202 通信経路
03 自律走行システム
030 自律走行制御装置
0301 自律動作制御装置
0302 記録装置
03021 動作制御論理の生成記録
0303 期待動作決定論理
0304 制約形式変換手段A
0305 制約形式変換手段B
031 撮像装置
032 車載センサ群
033 交通指示器・表示装置
034 駆動装置
Claims (6)
- 記憶装置に格納された動作制御論理に基づき、自システムの内部状態と、外界計測手段により取得される開放外部環境における外部要因に適応した動作を実現する動作制御論理を生成して駆動装置に対する開放外部環境における動作を指示する自律制御装置であって、
前記自律制御装置は、
前記外界計測手段から取得した外部要因と自システムの内部状態に対応した制約条件と、動作制御論理を逐次再構築する度に、自システムに求められている期待動作及び動作の健全性に関わる制約条件から、前記動作制御論理と制御論理を生成し、生成した当該動作制御論理と前記外界計測手段から取得した情報を前記記録装置に書き出し、前記制御論理に基づき自システムの動作を指示する自律動作制御装置と、
前記自律動作制御装置により前記期待動作及び動作の健全性の判断結果を元に、自システムに求められている期待動作を指示する期待動作決定部を備えること
を特徴とする自律制御装置。 - 請求項1に記載の自律制御装置において、
前記制約条件は、各制約条件の充足状態に応じて制約が緩和される緩和条件付きの制約条件であり、
前記自律動作制御装置は、前記期待動作の充足可能性を判定して、充足可能時及び充足不能時それぞれの場合の動作指示に対応した制御論理を生成する
ことを特徴とする自律制御装置。 - 請求項2に記載の自律制御装置において、
前記自律動作制御装置は、論理式又は前記論理式を含む時相論理の形式とした制約条件と許容緩和限界を含む制約形式を入力とし、
前記自律制御装置は、前記自システムの内部状態と開放外部環境における外部要因を前記制約形式に変換し、当該変換された制約形式のリストを生成する第1の変換手段と、
前記期待動作、及び前記健全動作を制約形式に変換し、当該変換された制約形式のリストを生成する第2の変換手段を備えることを特徴とする自律制御装置。 - 請求項3に記載の自律制御装置において、前記自律動作制御装置が前記健全動作に関する要件及び前記期待動作を共に実現する動作制御論理の存在を判定し、動作制御論理が存在する場合には、前記制御論理を逐次生成して前記期待動作を実現し、
前記動作制御論理が存在しない場合には、前記制約形式のリストの内、充足不能となった阻害要因を特定して、前記緩和条件の許容緩和限界を充足しながら、健全動作を回復するように制約形式のリストを選択して動作制御論理を生成することを特徴とする自律制御装置。 - 請求項2に記載の自律制御装置であって、
前記自律制御装置は、前記健全な期待動作を実現出来る場合には動作制御論理の内容を提示装置に提示し、前記健全な期待動作を実現できない場合には、阻害要因を提示装置に提示することを特徴とする自律制御装置。 - 請求項3に記載の自律制御装置を備えた移動体であって、前記期待動作は指定速度による走行指示であり、前記健全動作は前記移動体周囲の他の移動体又は障害物との接触を回避する動作であり、当該期待動作と健全動作を両立するように前記移動体の移動速度及び軌道を制御する機能を有することを特徴する移動体。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP13878317.0A EP2975481A4 (en) | 2013-03-11 | 2013-03-11 | Autonomous control device |
JP2015505094A JP6141964B2 (ja) | 2013-03-11 | 2013-03-11 | 自律制御装置 |
US14/650,111 US9983553B2 (en) | 2013-03-11 | 2013-03-11 | Autonomous control device |
PCT/JP2013/056559 WO2014141351A1 (ja) | 2013-03-11 | 2013-03-11 | 自律制御装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2013/056559 WO2014141351A1 (ja) | 2013-03-11 | 2013-03-11 | 自律制御装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014141351A1 true WO2014141351A1 (ja) | 2014-09-18 |
Family
ID=51536054
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/056559 WO2014141351A1 (ja) | 2013-03-11 | 2013-03-11 | 自律制御装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US9983553B2 (ja) |
EP (1) | EP2975481A4 (ja) |
JP (1) | JP6141964B2 (ja) |
WO (1) | WO2014141351A1 (ja) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016129067A1 (ja) * | 2015-02-12 | 2016-08-18 | 株式会社日立製作所 | 自律動作検証装置および自律システム |
CN106873596A (zh) * | 2017-03-22 | 2017-06-20 | 北京图森未来科技有限公司 | 一种车辆控制方法及装置 |
JP2017174208A (ja) * | 2016-03-24 | 2017-09-28 | トヨタ自動車株式会社 | 自動運転車両を用いたタクシーサービスシステム |
WO2021038826A1 (ja) * | 2019-08-30 | 2021-03-04 | 株式会社日立製作所 | 状態遷移モデル構築装置及び自律システム |
WO2022180788A1 (ja) | 2021-02-26 | 2022-09-01 | 日本電気株式会社 | 制約条件学習装置、制約条件学習方法及び記憶媒体 |
WO2022244143A1 (ja) * | 2021-05-19 | 2022-11-24 | 株式会社やまびこ | ロボット作業装置 |
WO2024034579A1 (ja) * | 2022-08-08 | 2024-02-15 | 川崎重工業株式会社 | 制御装置 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10241515B2 (en) * | 2015-05-29 | 2019-03-26 | Clearpath Robotics, Inc. | Method, system and apparatus for handling operational constraints for control of unmanned vehicles |
WO2018213931A1 (en) | 2017-05-25 | 2018-11-29 | Clearpath Robotics Inc. | Systems and methods for process tending with a robot arm |
US11191005B2 (en) | 2019-05-29 | 2021-11-30 | At&T Intellectual Property I, L.P. | Cyber control plane for universal physical space |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08221126A (ja) * | 1995-02-10 | 1996-08-30 | Yaskawa Electric Corp | 移動ロボット |
JP2008142841A (ja) * | 2006-12-11 | 2008-06-26 | Toyota Motor Corp | 移動ロボット |
JP2009080804A (ja) | 2007-09-03 | 2009-04-16 | Hitachi Industrial Equipment Systems Co Ltd | 自律移動ロボットシステム |
JP2009157735A (ja) | 2007-12-27 | 2009-07-16 | Toyota Motor Corp | 移動体、及び車両用警告装置 |
JP2012030665A (ja) | 2010-07-29 | 2012-02-16 | Toyota Motor Corp | 車両制御システム |
JP2012051560A (ja) | 2010-08-31 | 2012-03-15 | Atlas Elektronik Gmbh | 無人の水中航走体および無人の水中航走体を運転するための方法 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63196944A (ja) * | 1987-02-12 | 1988-08-15 | Hitachi Ltd | ル−ル検証方式 |
JP4107605B2 (ja) * | 2005-02-01 | 2008-06-25 | シャープ株式会社 | 移動体周辺監視装置、移動体周辺監視方法、制御プログラムおよび可読記録媒体 |
JP4670807B2 (ja) * | 2006-12-19 | 2011-04-13 | トヨタ自動車株式会社 | 移動経路作成方法、自律移動体および自律移動体制御システム |
US8428843B2 (en) | 2008-06-20 | 2013-04-23 | GM Global Technology Operations LLC | Method to adaptively control vehicle operation using an autonomic vehicle control system |
US8254670B2 (en) * | 2009-02-25 | 2012-08-28 | Toyota Motor Engineering & Manufacturing North America, Inc. | Self-learning object detection and classification systems and methods |
US8676466B2 (en) * | 2009-04-06 | 2014-03-18 | GM Global Technology Operations LLC | Fail-safe speed profiles for cooperative autonomous vehicles |
US8150621B1 (en) * | 2009-04-07 | 2012-04-03 | The United States of America as represeneted by the Secretary of the Navy | Command and control of autonomous surface vehicle |
US8595321B2 (en) | 2010-07-02 | 2013-11-26 | Nguyen Xuan Hoang | Supporting system for remote control |
-
2013
- 2013-03-11 JP JP2015505094A patent/JP6141964B2/ja active Active
- 2013-03-11 WO PCT/JP2013/056559 patent/WO2014141351A1/ja active Application Filing
- 2013-03-11 EP EP13878317.0A patent/EP2975481A4/en not_active Ceased
- 2013-03-11 US US14/650,111 patent/US9983553B2/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08221126A (ja) * | 1995-02-10 | 1996-08-30 | Yaskawa Electric Corp | 移動ロボット |
JP2008142841A (ja) * | 2006-12-11 | 2008-06-26 | Toyota Motor Corp | 移動ロボット |
JP2009080804A (ja) | 2007-09-03 | 2009-04-16 | Hitachi Industrial Equipment Systems Co Ltd | 自律移動ロボットシステム |
JP2009157735A (ja) | 2007-12-27 | 2009-07-16 | Toyota Motor Corp | 移動体、及び車両用警告装置 |
JP2012030665A (ja) | 2010-07-29 | 2012-02-16 | Toyota Motor Corp | 車両制御システム |
JP2012051560A (ja) | 2010-08-31 | 2012-03-15 | Atlas Elektronik Gmbh | 無人の水中航走体および無人の水中航走体を運転するための方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP2975481A4 |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016129067A1 (ja) * | 2015-02-12 | 2016-08-18 | 株式会社日立製作所 | 自律動作検証装置および自律システム |
JPWO2016129067A1 (ja) * | 2015-02-12 | 2017-10-26 | 株式会社日立製作所 | 自律動作検証装置および自律システム |
JP2017174208A (ja) * | 2016-03-24 | 2017-09-28 | トヨタ自動車株式会社 | 自動運転車両を用いたタクシーサービスシステム |
CN106873596A (zh) * | 2017-03-22 | 2017-06-20 | 北京图森未来科技有限公司 | 一种车辆控制方法及装置 |
US11500375B2 (en) | 2017-03-22 | 2022-11-15 | Beijing Tusen Weilai Technology Co., Ltd. | Method and apparatus for vehicle control |
US12024209B2 (en) | 2017-03-22 | 2024-07-02 | Beijing Tusen Weilai Technology Co., Ltd. | Method and apparatus for vehicle control |
WO2021038826A1 (ja) * | 2019-08-30 | 2021-03-04 | 株式会社日立製作所 | 状態遷移モデル構築装置及び自律システム |
WO2022180788A1 (ja) | 2021-02-26 | 2022-09-01 | 日本電気株式会社 | 制約条件学習装置、制約条件学習方法及び記憶媒体 |
WO2022244143A1 (ja) * | 2021-05-19 | 2022-11-24 | 株式会社やまびこ | ロボット作業装置 |
WO2024034579A1 (ja) * | 2022-08-08 | 2024-02-15 | 川崎重工業株式会社 | 制御装置 |
Also Published As
Publication number | Publication date |
---|---|
EP2975481A1 (en) | 2016-01-20 |
US20150309485A1 (en) | 2015-10-29 |
JP6141964B2 (ja) | 2017-06-07 |
JPWO2014141351A1 (ja) | 2017-02-16 |
EP2975481A4 (en) | 2017-01-18 |
US9983553B2 (en) | 2018-05-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2014141351A1 (ja) | 自律制御装置 | |
CN104108395B (zh) | 用于修改机动车辆的驾驶辅助***的配置的方法和装置 | |
US11288963B2 (en) | Autonomous vehicles featuring vehicle intention system | |
US10255168B2 (en) | Method and device for generating test cases for autonomous vehicles | |
US10571916B2 (en) | Control method for autonomous vehicles | |
US20220179418A1 (en) | Depart constraints implementation in autonomous vehicle routing | |
US20210173401A1 (en) | Human Supervision of an Automated Driving System | |
US20230234613A1 (en) | Testing and simulation in autonomous driving | |
CN111258217A (zh) | 实时对象行为预测 | |
Maierhofer et al. | Formalization of intersection traffic rules in temporal logic | |
US11628865B2 (en) | Method and system for behavioral cloning of autonomous driving policies for safe autonomous agents | |
KR20230160807A (ko) | 모바일 로봇 궤적 계획기들에 대한 성능 테스트 | |
CN113665593A (zh) | 一种车辆智能驾驶纵向控制方法、***及存储介质 | |
Sama et al. | Driving feature extraction and behavior classification using an autoencoder to reproduce the velocity styles of experts | |
US20230391357A1 (en) | Methods and apparatus for natural language based scenario discovery to train a machine learning model for a driving system | |
KR20210068449A (ko) | 자동차의 경로 및/또는 궤적을 계획하는 장치 | |
EP4397555A1 (en) | Software architecture platform and control method for autonomous vehicle, vehicle, and medium | |
WO2022072764A1 (en) | Metrics for evaluating autonomous vehicle performance | |
Noh et al. | Situation assessment and behavior decision for vehicle/driver cooperative driving in highway environments | |
Tong et al. | A search-based motion planner utilizing a monitoring functionality for initiating minimal risk maneuvers | |
CN116880462A (zh) | 自动驾驶模型、训练方法和自动驾驶方法和车辆 | |
KR20240019268A (ko) | 자율주행 차량 테스트를 위한 지원 도구 | |
Doss et al. | Predicting Desired Temporal Waypoints from Camera and Route Planner Images using End-To-Mid Imitation Learning | |
EP4332824A1 (en) | System and method for translating natural language traffic rules into formal logic for autonomous moving vehicles | |
EP4030408A1 (en) | Autonomous junction crossing of automated vehicle |
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: 13878317 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2015505094 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14650111 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2013878317 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |