DE112021007256T5 - BEHAVIOR DECISION APPARATUS, BEHAVIOR DECISION METHODS, BEHAVIOR DECISION PROGRAM, DESIGN ASSISTANCE DEVICE, DESIGN ASSISTANCE METHOD, AND DESIGN ASSISTANCE PROGRAM - Google Patents

Abstract

A rule system creation unit (212) constructs a logical system of a rule set, which is a set of traffic rules expressed by logical expressions. For each state transition in a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, a conversion unit (213) converts the state transition into a logical expression. A false transition detection unit (214) determines whether or not the logical expression of each state transition in the state transition graph contradicts the logical system of the rule set. A false transition removal unit (215) removes a state transition corresponding to each logical expression determined to be contradictory from the state transition graph as a false transition. A behavior decision unit (216) decides the behavior of the autonomous vehicle using the state transition graph after removing each false transition.

Description

TECHNICAL FIELD

The present disclosure relates to a technique for determining the behavior of an autonomous vehicle.

STATE OF THE ART

Conventionally, a behavioral decision-making technique for an autonomous vehicle has been proposed.

For example, Patent Literature 1 discloses a technique for calculating the next state to be assumed by one's own vehicle. With this technology, a condition of your own vehicle and an observed value from a sensor are first determined. The observed value of the sensor corresponds to the movement information of another vehicle. The next assumed state of the own vehicle is then calculated using a state transition graph of the own vehicle and a movement model of the other vehicle.

The state transition graph is a set of transitions between the states of your own vehicle. For example, part of the state transition graph indicates a transition from the "approaching the intersection" state to the "stopping before the intersection" state and specifies a condition for the "signal is red or yellow" transition.

REFERENCE LIST

PATENT LITERATURE

Patent literature 1: US 2019/0391580 A1

SUMMARY OF THE INVENTION

TECHNICAL PROBLEM

An autonomous vehicle must comply with traffic laws and regulations.

In traditional methods, a state transition graph must be designed that does not violate traffic laws and regulations. However, it is difficult for humans to fully understand the laws and regulations described comprehensively and in detail and to design a state transition graph consistent with the laws and regulations.

The present disclosure is intended to make it possible to determine the behavior of an autonomous vehicle while ensuring that it complies with traffic laws and regulations.

THE SOLUTION OF THE PROBLEM

• eine Regelsystem-Erstellungseinheit, der sich auf einen Regelsatz bezieht, der ein Satz von Verkehrsregeln ist, die durch logische Ausdrücke ausgedrückt werden, und um ein logisches System des Regelsatzes zu konstruieren;
• eine Umwandlungseinheit, der sich auf einen Zustandsübergangsgraphen bezieht, der eine Vielzahl von Zustandsübergängen für eine Vielzahl von Zuständen eines autonomen Fahrzeugs darstellt, und um den Zustandsübergang in einen logischen Ausdruck für jeden Zustandsübergang in dem Zustandsübergangsgraphen umzuwandeln;
• eine Falscher-Übergang-Erkennungseinheit, um für jeden Zustandsübergang in dem Zustandsübergangsgraphen festzustellen, ob der logische Ausdruck des Zustandsübergangs im Widerspruch zum logischen System des Regelsatzes steht oder nicht;
• eine Falscher-Übergang-Entfernungseinheit, um einen Zustandsübergang als falschen Übergang aus dem Zustandsübergangsgraphen zu entfernen, derjedem logischen Ausdruck entspricht, der als im Widerspruch zu dem logischen System des Regelsatzes stehend bestimmt wurde; und
• eine Verhaltensentscheidungseinheit, um über das Verhalten des autonomen Fahrzeugs unter Verwendung des Zustandsübergangsgraphen zu entscheiden, nachdem jeder falsche Übergang entfernt wurde.

Behavioral decision apparatus according to the present disclosure includes:

• a rule system creation unit, which refers to a rule set, which is a set of traffic rules expressed by logical expressions, and to construct a logical system of the rule set;
• a conversion unit relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and to convert the state transition into a logical expression for each state transition in the state transition graph;
• a false transition detection unit for determining, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set;
• a false transition removal unit for removing a state transition as a false transition from the state transition graph corresponding to any logical expression determined to conflict with the logical system of the ruleset; and
• a behavior decision unit to decide the behavior of the autonomous vehicle using the state transition graph after removing each false transition.

ADVANTAGEOUS EFFECTS OF THE INVENTION

According to the present disclosure, a state transition graph is modified based on a rule set, and the behavior of an autonomous vehicle is determined based on the modified state transition graph. This makes it possible to determine the behavior of the autonomous vehicle while ensuring compliance with traffic laws and regulations.

BRIEF DESCRIPTION OF DRAWINGS

• 1 is a configuration diagram of an autonomous vehicle 100 in Embodiment 1.
• 2 is a configuration diagram of a behavior decision device 200 in Embodiment 1.
• 3 is a functional configuration diagram of an autonomous driving system 110 in Embodiment 1.
• 4 is a flowchart of a behavioral decision process in Embodiment 1.
• 5 is a configuration diagram of a design assistant 300 in Embodiment 2.
• 6 is a functional configuration diagram of the design assistant 300 in Embodiment 2
• 7 is a flowchart of a design assistance process (transition) in Embodiment 2.
• 8th is a flowchart of the design assistance method (rule) in Embodiment 2.
• 9 is a hardware configuration diagram of the behavior decision device 200 in Embodiment 1.
• 10 is a hardware configuration diagram of the design assistant 300 in Embodiment 2.

DESCRIPTION OF EMBODIMENTS

In the embodiments and drawings, identical or corresponding elements are designated with the same reference numerals. The description of an element provided with the same reference numeral as an element already described is appropriately omitted or simplified. Arrows in diagrams primarily indicate data or processing flows.

Embodiment 1.

An embodiment for deciding on the behavior of an autonomous vehicle 100 is based on 1 until 4 described.

*** Description of the configuration ***

Based on 1 A configuration of the autonomous vehicle 100 will be described.

The autonomous vehicle 100 is a vehicle that has an autonomous driving function.

The autonomous vehicle 100 includes an autonomous driving system 110.

The autonomous driving system 110 is a system that realizes autonomous driving.

The autonomous driving system 110 includes a sensor group 111, an information acquisition device 112, a behavior decision device 200 and an autonomous drive device 113.

The sensor group 111 is one or more sensors installed in the autonomous vehicle 100.

Specifically, the sensor group 111 includes one or more sensors (a first sensor group) for detecting its own vehicle information and one or more sensors (a second sensor group) for detecting other vehicle information.

A specific example of the own vehicle information is position and speed information about the autonomous vehicle 100 (own vehicle). A concrete example of the first sensor group is a receiver and a speedometer of a satellite positioning system.

A specific example of the information about other vehicles is position information about each vehicle (other vehicle) that is within the radius of the autonomous vehicle 100. A specific example of a sensor in the second sensor group is a visible camera or a millimeter wave radar.

The information acquisition device 112 is a computer that uses the sensor group 111 to acquire its own vehicle information and the information of other vehicles.

For example, the receiver of the satellite positioning system measures a position of the autonomous vehicle 100, and the information acquisition device 112 acquires the position information about the autonomous vehicle 100 from the receiver of the satellite positioning system.

The visible camera, for example, records the surroundings of the autonomous vehicle 100 and generates image data. The image data represents an image showing the periphery of the autonomous vehicle 100. Then, the information acquisition device 112 acquires the image data from the visible camera, calculates a relative position of each other vehicle with respect to the autonomous vehicle 100 by processing the image data, and generates the position information indicating the relative position of each other vehicle.

For example, the millimeter-wave radar calculates a relative direction and a relative distance of each other vehicle with respect to the autonomous vehicle 100. Then, the information acquisition device 112 detects from the millimeter-wave radar the relative direction and relative distance of each other vehicle, calculates the relative position of each other vehicle with respect to to the autonomous vehicle 100 based on the relative direction and distance of each other vehicle and generates the position information indicating the relative position of each other vehicle.

The behavior decision device 200 is a computer that decides the behavior of the autonomous vehicle 100 based on its own vehicle information and the other vehicle information.

For example, the behavior of the autonomous vehicle 100 is such as acceleration and steering.

The autonomous driving device 113 is a computer that performs autonomous driving by controlling the autonomous vehicle 100 according to the behavior determined by the behavior decision device 200.

Based on 2 A configuration of the behavior decision device 200 will be described.

The behavior decision device 200 is a computer that includes hardware components such as a processor 201, a memory 202, an auxiliary storage device 203 and an input/output interface 204. These pieces of hardware are connected to each other by signal lines.

The processor 201 is an IC that performs operational processing and controls other parts of the hardware. The processor 201 is, for example, a CPU.

IC is an abbreviation for Integrated Circuit.

CPU is an abbreviation for Central Processing Unit.

Memory 202 is a volatile or non-volatile memory device. The memory 202 is also referred to as a main storage device or a main memory. For example, memory 202 is RAM. Data stored in the memory 202 is backed up in the auxiliary storage device 203 as needed.

RAM is an abbreviation for Random Access Memory.

The auxiliary storage device 203 is a non-volatile storage device. For example, the auxiliary storage device 203 is a ROM, an HDD, or a flash memory. Data stored in the auxiliary storage device 203 is loaded into the memory 202 as needed.

RAM is an abbreviation for Read Only Memory.

HDD is the abbreviation for Hard Disk Drive.

The input/output interface 204 is an interface for input/output of data. The input/output interfaces 204 are, for example, different ports.

The behavior decision device 200 includes elements such as an update unit 211, a control system creation unit 212, a conversion unit 213, a false transition detection unit 214, a false transition removal unit 215, and a behavior decision unit 216. These elements are implemented by software.

The auxiliary storage device 203 stores a behavior decision program to cause a computer to function as the update unit 211, the control system creation unit 212, the conversion unit 213, the false transition detection unit 214, the false transition removal unit 215 and the behavior decision unit 216. The behavioral decision program is loaded into memory 202 and executed by processor 201.

Further, the auxiliary storage device 203 stores an OS. At least a portion of the OS is loaded into memory 202 and executed by processor 201.

The processor 201 executes the behavior decision program during execution of the operating system.

OS is an abbreviation for Operating System.

Input and output data of the behavioral decision program are stored in a storage unit 290.

The memory 202 functions as the storage unit 290. However, a storage device such as the auxiliary storage device 203, a register in the processor 201, and a cache memory in the processor 201 may operate as the storage unit 290 instead of the memory 202 or together with the memory 202.

The behavior decision device 200 may include a plurality of processors as an alternative to the processor 201.

The behavioral decision program may be recorded (stored) in a computer-readable format on a non-volatile recording medium such as an optical disk or flash memory.

3 shows a functional configuration of the autonomous driving system 110. An arrow in the diagram indicates a data flow or a processing flow.

The operation of the elements (221 to 226) and the contents of the data (291 and 292) in the behavior decision device 200 will be described below.

*** Description of how it works ***

A method of operating the behavior decision device 200 corresponds to a behavior decision method. In addition, the method for operating the behavior decision device 200 corresponds to a method for processing by the behavior decision program.

Based on 4 the behavioral decision-making process is described.

In step S110, the rule system creation unit 212 refers to a rule set 292 and constructs a logical system of the rule set 292.

The rule set 292 is a set of traffic rules expressed by logical expressions and stored in the storage unit 290.

Traffic rules are legal rules that must be observed when driving. However, traffic rules can also be other than legal rules, as long as they must be observed.

For example, traffic rules are expressed through logical expressions as follows.

Let “a” be the statement symbol indicating that “the traffic light is currently red.”

Let “b” be the statement symbol indicating that “the traffic light is currently yellow”.

Let C'' be the statement symbol that indicates that “an intersection is being entered”.

The traffic rule that says not to enter an intersection when the signal is red or yellow is expressed by the following logical expression. “|” means a logical sum. “^” means negation. $$\left(\text{a}|\text{b}\right)\to \text{c}$$

However, the traffic rule can also be expressed using another expression method as long as a logical system can be constructed.

For example, the logical system of rule set 292 is structured as follows.

The rule set 292 consists of a traffic rule A, a traffic rule B and a traffic rule C. In this case, the logical system of the rule set 292 is expressed as follows.

ABC

However, the logic system of rule set 292 can be constructed using a different construction method as long as it is possible to detect a false transition, which will be described below.

In step S120, the updating unit 211 updates a state transition graph 291 based on the own vehicle information and the information about other vehicles.

The state transition graph 291 is a data graph representing a plurality of state transitions for a plurality of states of the autonomous vehicle 100 and is stored in the storage unit 290.

The state transition means a transition between states. For example, in state transition graph 291, each state transition is represented by two nodes and one edge. Each node represents a state. The edge represents a direction of transition and a condition of transition.

In particular, the update unit 211 calculates various information on the Based on the own vehicle information and the other vehicle information, and updates the state transition graph 291 based on the calculated various information.

The state transition graph 291 can be updated using any method.

In addition, step S120 can be omitted. That is, the state transition graph 291 may not be updated based on the information about the own vehicle and the other vehicles.

In step S130, the conversion unit 213 selects an unselected state transition from the state transition graph 291.

Steps S140 to S160 are carried out for the state transition selected in step S130.

In step S140, the conversion unit 213 converts the state transition into a logical expression.

The state transition is represented, for example, by a logical expression as follows.

Der Zustandsübergang kann jedoch auch mit einem anderen Umwandlungsverfahren umgewandelt werden, solange es möglich ist, den unten zu beschreibenden falschen Übergang zu erkennen.A state of a transition source is “s1”. A state of a transition target is “s2”. A condition for the transition is “c1” and “c2, or^c3”. In this case, the state transition is expressed by the following logical expression. $$\left(\text{s1\&}\left(\text{c1\&}\left(\text{c2}|\text{c3}\right)\right)\right)\to \text{s2}$$

However, the state transition can also be converted using another conversion method as long as it is possible to detect the false transition to be described below.

In step S150, the false transition detection unit 214 determines whether or not the logical expression of the state transition contradicts the logical system of the rule set 292.

If the logical expression of the state transition contradicts the logical system of Rule Set 292, the state transition is considered a violation of Rule Set 292.

For example, the false transition detection unit 214 determines whether a logical expression A in a logical system L is true or false by solving a satisfiability problem of L&P.

However, the false transition detection unit 214 may also use another method for determination.

If the logical expression of the state transition contradicts the logical system of the rule set 292, processing proceeds to step S160.

If the logical expression of the state transition does not contradict the logical system of the rule set 292, processing proceeds to step S170.

A state transition that corresponds to the logical expression determined to contradict the logical system of Rule 292 is called a “false transition.”

In step S160, the false transition removal unit 215 temporarily removes the false transition from the state transition graph 291.

The temporal elimination of the incorrect transition means that the incorrect transition is no longer taken into account when deciding on the next behavior of the autonomous vehicle 100.

In step S170, the conversion unit 213 determines whether or not there is an unselected state transition in the state transition graph 291.

If the unselected state transition is present in the state transition graph 291, processing proceeds to step S130.

If there is no unselected state transition in the state transition graph 291, processing proceeds to step S180.

In step S180, the behavior decision unit 216 decides the next behavior of the autonomous vehicle 100 using the state transition graph 291 after removing each false transition.

Specifically, the behavior decision unit 216 decides the next behavior of the autonomous vehicle 100 as follows based on the state transition graph 291, the own vehicle information and the other vehicle information.

First, the behavioral decision unit 216 selects from the state transition graph phen 291 selects a state (a transition source state) that is consistent with the current state of the autonomous vehicle 100, and extracts from the state transition graph 291 one or more states (states of transition destinations) to which the selected state transition transitions.

Next, the behavior decision unit 216 decides, from one or more states of the transition goals, a state of a transition goal as the next state of the autonomous vehicle 100 based on the own vehicle information, the other vehicle information and a motion model of the autonomous vehicle 100. For example, the next state of the autonomous vehicle 100 determined according to the method described in Patent Document 1.

Then, the behavior decision unit 216 decides the next behavior of the autonomous vehicle 100 based on the current state of the autonomous vehicle 100 and the next state of the autonomous vehicle 100. The next behavior of the autonomous vehicle 100 is determined according to any method.

*** Description of the operating examples ***

The rule system creation unit 212 may incorporate dynamic environmental information detected by the sensor group 111 into the logical system of the rules set 292, e.g. weather information that indicates the weather or time information that indicates the time. The sensor group 111 includes, for example, a receiver that receives the weather information and a receiver that receives the time information.

The dynamic environmental information is reflected in the traffic rules as dynamic conditions. The time information is reflected, for example, in traffic rules, such as “no passage after xx o’clock”. In addition, the weather information is reflected in the traffic rules, e.g. E.g. “you have to drive yy in fog”.

The rule system creation unit 212 may add a logical expression to the logic system of the rules set 292 indicating that two or more states of the autonomous vehicle 100 are not simultaneously true.

***Effects of Embodiment 1***

The behavior decision device 200 can ensure compliance with the rule set 292 as follows.

The behavior decision device 200 decides the behavior of the autonomous vehicle 100 using the state transition graph 291, which consists only of state transitions that do not violate the rule set 292. This means that the behavior decision device 200 is able to decide the behavior that does not violate the rule set 292.

The behavior decision device 200 modifies the state transition graph 291 so that the rule set 292 is satisfied by logical inference using the rule set 292.

To decide on the behavior of the autonomous vehicle 100, the behavior decision algorithm described in Patent Document 1 can be used unchanged. The behavioral decision algorithm described in Patent Literature 1 is closely linked to the motion model.

In general, a number of legal rules are described in detail and in detail. Therefore, it is difficult to capture all legal rules and manually build a state transition graph that is consistent with all legal rules. In Embodiment 1, the state transition graph 291 is modified based on the ruleset 292 so that a creator can create the state transition graph 291 without knowing the ruleset 292.

The expression and application of the rules of law must be strict and consistent with logical conclusions and expression using a logical expression. In addition, it is necessary not only to validate the individual legal rules, but also to ensure composite rules that can be established by recursively combining legal rules. Therefore, the determination based on the logical conclusion as in Embodiment 1 is appropriate to determine the behavior of the autonomous vehicle 100.

Embodiment 2.

With regard to an embodiment in which the design of the state transition graph 291 and the ruleset 292 is supported, who the differences from embodiment 1 are mainly based on the 5 until 8th described.

*** Description of the configuration ***

Based on 5 A configuration of a design assistance device 300 is described.

The design assistance device 300 is a computer that includes hardware parts such as a processor 301, a memory 302, an auxiliary storage device 303, a communication device 304 and an input/output interface 305. These pieces of hardware are connected to each other by signal lines.

The processor 301 is an IC that performs operational processing and controls other parts of the hardware. The processor 301 is, for example, a CPU.

Memory 302 is a volatile or non-volatile memory device. The memory 302 is also referred to as a main storage device or a main memory. For example, memory 302 is RAM. Data stored in the memory 302 is backed up in the auxiliary storage device 303 as needed.

The auxiliary storage device 303 is a non-volatile storage device. For example, the auxiliary storage device 303 is a ROM, an HDD, or a flash memory. Data stored in the auxiliary storage device 303 is loaded into the memory 302 as needed.

The communication device 304 is a receiver and a transmitter. For example, the communication device 304 is a communication chip or NIC. The communication of the design assistance device 300 takes place via the communication device 304.

The input/output interface 305 is a port to which an input device and an output device are connected. For example, the input/output interface 305 is a USB port, the input device is a keyboard and mouse, and the output device is a display. The input/output of the design assistance device 300 takes place via the input/output interface 305.

The design assistance device 300 includes elements such as a rule system creation unit 311, a conversion unit 312, a false transition detection unit 313, a transition system creation unit 314, a non-derived rule detection unit 315 and a display unit 316. These elements are implemented by software.

The auxiliary storage device 303 stores a design assistance program to cause a computer to act as the rule system creation unit 311, the conversion unit 312, the false transition detection unit 313, the transition system creation unit 314, the non-derived rule detection unit 315 and the display unit 316 act. The design assistance program is loaded into memory 302 and executed by processor 301.

Further, the auxiliary storage device 303 stores an OS. At least a portion of the OS is loaded into memory 302 and executed by processor 301.

The processor 301 executes the design assistance program while executing the operating system.

Input and output data of the design assistance program are stored in a storage unit 390.

The memory 302 functions as the storage unit 390. However, a storage device such as the auxiliary storage device 303, a register in the processor 301 and a cache memory in the processor 301, may function as the storage unit 390 instead of the memory 302 or together with the memory 302 .

The design assistance device 300 may contain a variety of processors as an alternative to the processor 301.

The design assistance program may be recorded (stored) in a computer-readable format on a non-volatile recording medium such as an optical disk or flash memory.

6 shows a functional configuration of the design assistance device 300. An arrow in the diagram indicates a data flow or a processing flow.

The state transition graph 291 and the rule set 292 are stored in the storage unit 390.

The functionality of the individual elements of the design assistance device 300 is described below.

*** Description of how it works ***

A method for operating the design assistance device 300 corresponds to a design assistance method. In addition, the method of operating the design assistance device 300 corresponds to a method of processing by the design assistance program.

Based on 7 The design assistance process (transition) is described.

The design assistance method (transition) is a method for displaying an incorrect transition.

In step S301, the rule system creation unit 311 constructs the logical system of the rule set 292.

A construction method is the same as the method in step S110 of Embodiment 1.

In step S302, the conversion unit 312 selects an unselected state transition from the state transition graph 291.

Steps S303 to S305 are carried out for the state transition selected in step S302.

In step S303, the conversion unit 312 converts the state transition into a logical expression.

A conversion process is the same as the process in step S140 of Embodiment 1.

In step S304, the false transition detection unit 313 determines whether the logical expression of the state transition contradicts the logical system of the rule set 292 or not.

A determination process is the same as the process in step S150 of Embodiment 1.

If the logical expression of the state transition contradicts the logical system of the rule set 292, processing proceeds to step S305.

If the logical expression of the state transition does not contradict the logical system of the rule set 292, processing continues to step S306.

A state transition that corresponds to the logical expression that contradicts the logical system of Rule 292 is called a “false transition”.

In step S305, the display unit 316 displays the incorrect transition on a display.

The wrong transition is equivalent to a state transition that violates the laws and regulations.

In step S306, the conversion unit 312 determines whether or not there is an unselected state transition in the state transition graph 291.

If the unselected state transition is present in the state transition graph 291, processing proceeds to step S302.

If there is no unselected state transition in the state transition graph 291, processing ends.

Based on 8th The design assistance process (rule) is described.

The design assistance process (rule) is a process for displaying a rule to be described below.

In step S311, the conversion unit 312 converts each state transition in the state transition graph 291 into a logical expression.

A conversion process is the same as the process in step S140 of Embodiment 1.

If each state transition in the state transition graph 291 has already been converted into a logical expression, step S311 is unnecessary.

In step S312, the transition system creation unit 314 constructs a logical system from a plurality of logical expressions for a plurality of logical expressions corresponding to a plurality of state transitions in the state transition graph 291.

The logical system to be constructed is referred to as the state transition graph logical system 291.

A method for constructing the logical system of the state transition graph 291 is the same as the procedure for constructing the logical system of rule set 292.

In step S313, the non-derived rule recognition unit 315 selects an unselected traffic rule from the rule set 292.

Steps S314 and S315 are carried out for the traffic rule selected in step S313.

In step S314, the non-derived rule recognition unit 315 determines whether or not a logical expression of the traffic rule is derived from the logical system of the state transition graph 291.

If the logical expression of the traffic rule is not derived from the logical system of the state transition graph 291, it is considered that there is a traffic rule that is not expressed in the state transition graph 291.

For example, the non-derivative rule recognizer 315 determines whether or not a logical expression A in a logical system L can be derived by solving satisfiability problems of L&P and ^(L&P).

However, the non-derived rule recognizer 315 may also use another method for determination.

If the logical expression of the traffic rule is derived from the logical system of the state transition graph 291, processing proceeds to step S316.

If the logical expression of the traffic rule is not derived from the logical system of the state transition graph 291, processing continues to step S315.

A traffic rule corresponding to the logical expression that is not derived from the logical system of the state transition graph 291 is called a “non-derived rule”.

In step S315, the display unit 316 displays the non-derived rule on a display.

The non-derived rule corresponds to a traffic rule that is not guaranteed in the state transition graph 291.

In step S316, the non-derived rule recognition unit 315 determines whether or not an unselected traffic rule exists in the rule set 292.

If the unselected traffic rule is present in the rule set 292, processing continues to step S313.

If there is no unselected traffic rule in rule set 292, processing ends.

*** Description of the operating examples ***

The rule system creation unit 311 may add a logical expression to the logic system of the rule set 292 indicating that two or more states of the autonomous vehicle 100 cannot be true at the same time.

The transition system creation unit 314 may add to the logical system of the state transition graph 291 the logical expression indicating that two or more states of the autonomous vehicle 100 cannot be true at the same time.

***Effects of Embodiment 2***

Embodiment 2 may be used by a designer of the state transition graph 291 to determine whether the state transition graph 291 is correct or incorrect and to redesign the state transition graph 291 if necessary.

Embodiment 2 may be used by a designer of the ruleset 292 to determine whether the ruleset 292 is correct or incorrect and to redesign the ruleset 292 if necessary.

The design assistant 300 displays in the state transition graph 291 a state transition (false transition) that violates the rule set 292 and a traffic rule (non-derived rule) that is not ensured. Displaying the assumed rule informs a designer of a possible missing state transition.

Therefore, the designer can design the state transition graph 291 to comply with the rule set 292.

*** Addition to embodiments ***

Based on 9 A hardware configuration of the behavior decision device 200 is described.

The behavior decision device 200 includes a processing circuit 209.

The processing circuit 209 is hardware that realizes the update unit 211, the control system creation unit 212, the conversion unit 213, the false transition detection unit 214, the false transition removal unit 215 and the behavior decision unit 216.

Processing circuitry 209 may be dedicated hardware or may be processor 201 that executes programs stored in memory 202.

If the processing circuit 209 is the dedicated hardware, the processing circuit 209 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

ASIC is an abbreviation for Application Specific Integrated Circuit.

FPGA is an abbreviation for Field Programmable Gate Array.

The behavior decision device 200 may also include a plurality of processing circuits as an alternative to the processing circuits 209.

In the processing circuit 209, some functions can be implemented by special hardware and the remaining functions by software or firmware.

As described above, the functions of the behavior decision device 200 may be implemented by hardware, software, firmware, or a combination thereof.

Based on 10 A hardware configuration of the design assistance device 300 is described.

The design assistance device 300 includes a processing circuit 309.

The processing circuit 309 is hardware that includes the rule system creation unit 311, the conversion unit 312, the false transition detection unit 313, the transition system creation unit 314, the non-derived rule detection unit 315, and the display unit 316.

Processing circuitry 309 may be dedicated hardware or may be processor 301 that executes programs stored in memory 302.

If the processing circuit 309 is the dedicated hardware, the processing circuit 309 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

The design assistance device 300 may include a variety of processing circuits as an alternative to the processing circuits 309.

In the processing circuit 309, some functions may be implemented by special hardware and the remaining functions may be implemented by software or firmware.

As described above, the functions of the design assistance device 300 may be implemented by hardware, software, firmware, or a combination thereof.

Each of the embodiments is an example of a preferred embodiment and is not intended to limit the technical scope of the present disclosure. Each of the exemplary embodiments may also be only partially implemented or may be implemented in combination with another exemplary embodiment. The processes described in the flowcharts or similar can be changed as necessary.

Each “entity” that is an element of the behavioral decision device 200 and the design assistance device 300 can be interpreted as a “process”, “step”, “circuit” or “circuit”.

REFERENCE SYMBOL LIST

100: autonomous vehicle; 110: autonomous driving system; 111: sensor group; 112: information acquisition device; 113: autonomous drive device; 200: behavioral decision device; 201: processor; 202: memory; 203: auxiliary storage device; 204: input/output interface; 209: processing circuit; 211: update unit; 212: Rule system creation unit; 213: conversion unit; 214: False transition detection unit; 215: False Transition Distance Unit; 216: Behavioral Decision Unit; 290: storage unit; 291: state transition graph; 292: rule set; 300: Design Assistance Facility; 301: processor; 302: memory; 303: auxiliary storage device; 304: communication device; 305: input/output interface; 309: processing circuit; 311: Rules system creation unit; 312: conversion unit; 313: False transition detection unit; 314: Transition system creation unit; 315: Non-derived rule recognition unit; 316: display unit; 390: storage unit

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2019/0391580 A1 [0005]

Claims (15)

Behavioral decision device comprising: a rule system creation unit that refers to a rule set, which is a set of traffic rules expressed by logical expressions, and to construct a logical system of the rule set; a conversion unit relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and converting the state transition into a logical expression for each state transition in the state transition graph; a false transition detection unit for determining, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set; a false transition removal unit for removing a state transition as a false transition from the state transition graph corresponding to each logical expression determined to conflict with the logical system of the ruleset; and a behavior decision unit to decide the behavior of the autonomous vehicle using the state transition graph after removing each false transition. Behavioral decision device Claim 1 , wherein the rule system creation unit reflects dynamic environmental information detected by a sensor installed on the autonomous vehicle to the logic system of the ruleset. Behavioral decision device Claim 1 or Claim 2 , wherein the rule system creation unit adds to the logical system of the ruleset a logical expression indicating that two or more states of the autonomous vehicle are not true at the same time. Behavioral decision-making process, comprising: Referring to a rule set, which is a set of traffic rules expressed by logical expressions, and constructing a logical system of the rule set; referring to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle and converting the state transition into a logical expression for each state transition in the state transition graph; determining, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set; removing a state transition as a false transition from the state transition graph corresponding to any logical expression determined to conflict with the logical system of the ruleset; and Determine the behavior of the autonomous vehicle using the state transition graph after each false transition has been removed. Behavioral decision program that causes a computer to execute: a rule system creation process that refers to a rule set, which is a set of traffic rules expressed by logical expressions, and to construct a logical system of the rule set; a conversion process relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and to convert the state transition into a logical expression for each state transition in the state transition graph; a false transition detection process to determine, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set; a false transition removal process to remove a state transition as a false transition from the state transition graph corresponding to each logical expression determined to conflict with the logical system of the ruleset; and a behavioral decision process to determine the behavior of the autonomous vehicle using the state transition graph after removing each false transition. A design assistance device comprising: a conversion unit relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and converting the state transition into a logical expression for each state transition in the state transition graph; a transition system creation unit for constructing, as a logical system of the state transition graph, a logical system from a plurality of logical expressions corresponding to the plurality of state transitions; a recognition unit for non-derived rules, to determine, for each traffic rule in a rule set that is a set of traffic rules expressed by logical expressions, whether or not the logical expression of the traffic rule is derived from the logical system of the state transition graph; and a display unit for displaying a traffic rule corresponding to a logical expression that cannot be derived from the logical system of the state transition graph. Design assistance facility Claim 6 , wherein the transition system creation unit adds to the logical system of the state transition graph a logical expression indicating that two or more states of the autonomous vehicle are not simultaneously true. Design assistance facility Claim 6 or Claim 7 , comprising: a rule system creation unit relating to a rule set, which is a set of traffic rules expressed by logical expressions, and for constructing a logical system of the rule set; and a false transition detection unit for determining, for each state transition in the state transition graph, whether or not the state transition contradicts the rule set using the logical expression of the state transition and the logical system of the rule set, the display unit indicating a state transition , which corresponds to any logical expression determined to contradict the logical system of the rule set. Design assistance facility Claim 8 , wherein the rule system creation unit adds to the logical system of the ruleset a logical expression indicating that two or more states of the autonomous vehicle are not true at the same time. Design assistance process, comprising: referring to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle and converting the state transition into a logical expression for each state transition in the state transition graph; constructing a logical system from a plurality of logical expressions corresponding to the plurality of state transitions as a logical system of the state transition graph; determining, for each traffic rule in a rule set that is a set of traffic rules expressed by logical expressions, whether or not the logical expression of the traffic rule is derived from the logical system of the state transition graph; and Display a traffic rule that corresponds to a logical expression that cannot be derived from the logical system of the state transition graph. Design assistant program that causes a computer to execute: a conversion process relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and to convert the state transition into a logical expression for each state transition in the state transition graph; a transition system creation process for constructing, as a logical system of the state transition graph, a logical system from a plurality of logical expressions corresponding to the plurality of state transitions; a non-derived rule recognition process to determine, for each traffic rule in a rule set, which is a set of traffic rules expressed by logical expressions, whether the logical expression of the traffic rule is derived from the logical system of the state transition graph or not; and a display process to display a traffic rule that corresponds to a logical expression that has been determined not to be derived from the logical system of the state transition graph. Design assistance device, comprising: a rule system creation unit that refers to a rule set, which is a set of traffic rules expressed by logical expressions, and to construct a logical system of the rule set; a conversion unit relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and converting the state transition into a logical expression for each state transition in the state transition graph; a false transition detection unit for determining, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set; and a display unit for displaying a state transition corresponding to each logical expression determined to conflict with the logical system of the rule set. Design assistance facility Claim 12 , wherein the rule system creation unit adds to the logical system of the ruleset a logical expression indicating that two or more states of the autonomous vehicle are not true at the same time. Design assistance process, comprising: Referring to a rule set, which is a set of traffic rules expressed by logical expressions, and constructing a logical system of the rule set; referring to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle and converting the state transition into a logical expression for each state transition in the state transition graph; determining, for each state transition in the state transition graph, whether or not the logical expression of the state transition conflicts with the logical system of the rule set; and Display a state transition for any logical expression that is determined to conflict with the ruleset's logical system. Design assistant program that causes a computer to execute: a rule system creation process that refers to a rule set, which is a set of traffic rules expressed by logical expressions, and to construct a logical system of the rule set; a conversion process relating to a state transition graph representing a plurality of state transitions for a plurality of states of an autonomous vehicle, and to convert the state transition into a logical expression for each state transition in the state transition graph; a false transition detection process to determine, for each state transition in the state transition graph, whether or not the logical expression of the state transition contradicts the logical system of the rule set; and a display process to display a state transition corresponding to each logical expression determined to conflict with the logical system of the ruleset.

Images