JPH07225612A - Method and device for path search having time base put in search space - Google Patents

Abstract

PURPOSE:To inputs the knowledge of an event which changes with time, to accurately judge the possibility of arrival, and search for the least-cost path by adding the time base to two-dimensional or three-dimensional search space bases, and rewriting the event which changes with time into an event of the space bases and performing processing. CONSTITUTION:A data base part 1 stores the timetable of a map and a moving body and the knowledge of the kinetic law of the moving body which is fixed to some extent, and a simulation part 2 performs simulation, event by event, on the basis of information from the data base part 1. A cost calculation part 3 integrates the influence of each event to calculate the cost in a search space, and converts a grating and a three-dimensional grating map wherein a probability distribution is described into a grating and a three-dimensional grating map described with the cost. A path search part 4 after finding several candidates for a path on the basis of the grating and three-dimensional grating maps described with the cost determines the path which is minimum in cost total and outputs the optimum path.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for searching a route of a mobile robot. The mobile robots that are currently in practical use, like parts transportation vehicles in factories, automatically operate a certain course guided by a line drawn on the floor in a well-controlled environment with a certain schedule. The present invention is directed to a route searching method in the above-mentioned autonomous mobile robot. , And devices.

[0002]

2. Description of the Related Art FIG. 14 is a diagram showing a configuration example of a conventional mobile robot, in which a thick frame shows the entire configuration and a dotted line frame shows a configuration example of a route generation unit.

The mobile robot is equipped with a camera for observing environmental conditions and sensors 13a and 13b such as ultrasonic sensors and touch sensors. When performing remote control, a human moves the mobile robot based on information from those sensors 13a and 13b.

In the case of performing autonomous movement instead of the above remote control, the route generation unit 14 is based on the map of the environment in which the mobile robot moves around in advance.
At, a route plan is created and a movement operation based on the route plan is executed. Specifically, based on the route information determined by the route generation unit 14, the central control unit (10)
Starts the drive control unit 11, and the drive unit 12
The route determined above is moved.

The route planning method is performed as follows. In other words, the environment map that describes the existence of obstacles such as walls and immovable objects and the time required to pass that location by a unit distance into costs is described as {cost for each grid or cubic grid. Based on the described cost map},
Calculate costs such as the total time required for several routes,
Various routes satisfying constraints such as reachability are found, and the route with the lowest cost is determined from them. When actually moving around, there are obstacles that are not on the map and moving objects such as humans and other mobile robots. Avoiding collisions with those that are not in the above-mentioned route planning is performed by the central control unit.
The operation in 10 is performed as an exception process, and the collision avoidance operation, which is determined separately from the movement operation according to the route plan, such as the detour operation, is executed. If the detour operation is not possible, the process returns to the movement start point, another route is determined, and the movement is performed.

[0006]

As described above, conventionally, when planning a route, the events in the map of the environment in which the mobile robot moves around are always time-invariant and there is no omission in the obtained information. , Is required to be accurate.

Therefore, knowledge of time-varying events such as doors that open and close at regular intervals and moving objects that move at a constant speed is used for the route planning, specifically, based on the cost map. There is a problem that when finding the candidate of and determining the route with the lowest cost among the routes, it cannot be incorporated in the route plan. Therefore, when there are such time-changing events, there is a problem in that the possibility of arrival is erroneously determined, or the cost cannot be reduced to a true minimum.

Further, as described above, the collision avoidance action is performed separately from the movement on the set route (that is, exception processing), so that the adjustment action between the two types of actions becomes complicated. there were. For example, when considering the route return after the collision avoidance operation, there is a problem that it is necessary to recognize the current position and direction of the mobile robot by new sensing and set a new route for returning to the target route. .

[0009] Here, if it is possible to incorporate them into a path plan in order to cope with a moving object whose motion can be predicted sufficiently accurately, it is possible to perform a more efficient movement than a movement operation by collision avoidance as a simple exception process. It should be.

In view of the above-mentioned conventional drawbacks, the present invention incorporates knowledge of a time-varying event into a route search in a route search method and apparatus for a mobile robot, accurately determines the possibility of reaching, and minimizes the cost. It is an object of the present invention to provide a route search method and device for searching the route.

[0011]

FIG. 1 is a block diagram showing the principle of the present invention. The above problems can be solved by a route search method and apparatus that incorporates a time axis in the search space configured as described below.

[0012] In the two-dimensional or three-dimensional search space used for the route search of the mobile robot, which describes the map and the information of the moving object, several route candidates are found to minimize the cost. Is a route search method for determining the route of the mobile robot and searching for a route along which the mobile robot moves. The route is changed by adding a time axis to the two-dimensional or three-dimensional search space axis. The event is configured to be processed by replacing it with the event on the spatial axis.

That is, in FIG. 1, the database unit 1 stores a certain degree of fixed knowledge such as a map, a timetable of a moving object, and a law of motion of a moving object. A simulation is performed for each event based on the information in. Specifically, when the mobile robot passes through, a timetable of the moving object, a grid describing a probability distribution according to the motion law of the moving object, and a cubic grid map are generated. At this time, the interaction between the events occurring in the mobile robot environment complicates the processing and is not taken into consideration. If an accurate prediction cannot be made, the prediction is made in the form of the distribution of the occurrence probability of the phenomenon. The cost calculation unit 3 calculates the cost in the search space by combining the influences of each event, and the grid that describes the above probability distribution, the cubic grid map, the grid that describes the cost,
Convert to cubic grid map.

The route search unit 4 uses a normal route search method based on the lattice and cubic lattice map described above.
For example, the maze method, which is often used in the field of CAD, is used to find several route candidates, then the route with the lowest total cost is determined, and the optimal route is output.

[0015]

First, the action will be considered in the case where the mobile robot has complete knowledge of all the events in the environment in which it moves around and the action of the moving object is accurately predicted.
In the coordinate axes of the search space, including events that do not change over time,
Comparing the search space with the addition of the time coordinate axis and the search space with only the space axis that includes only the events that do not change with time, the route search can be treated in the same way except for the following points. .

That is, since the time cannot be reversed and the moving speed of the moving object has the maximum limit, the direction of route setting is limited in the search space including the time axis. The reaching target point is determined by a combination of the destination point and the time limit.

By carrying out the route planning in this way, the collision avoidance behavior with the moving object can be processed in a unified manner with the route planning in the obstacle which does not change with time. In this method, since collision avoidance can be incorporated at the time of route planning, efficient movement can be performed, and there is an effect that it is not necessary to rebuild the route plan after collision avoidance.

However, in reality, accurate future prediction is impossible. Therefore, in the present invention, the uncertainties of those predictions are treated as the existence probability distribution. The cost used for route planning is calculated by combining the existence probability of the above-mentioned event and the influence of the degree of danger.

Regarding the future prediction, the most difficult problem is
It is the interaction between the events that occur in the environment. Predicting the interaction requires a huge amount of calculation as compared with performing a single prediction, so that interactions other than those having a significant influence are ignored. Furthermore, the interaction between the mobile robot and the events occurring in the environment causes the problem that the content of the search space changes during the route search.

Therefore, in the present invention, the route search method is premised on that the contents of the search space are fixed. Therefore, when the contents of the search space change, the route search is performed. Can't run. In the environment currently envisioned, the direction of interaction may be cooperative with respect to the mobile robot, such as avoiding a collision. In other words, if the interaction between the mobile robot and the events that occur in the environment is not taken into consideration, the efficiency will be poor, but it should not be dangerous. In consideration of this point, in the route search method and apparatus according to the present invention, the route search is performed without considering the interaction between the mobile robot and the event occurring in the environment.

[0021]

Embodiments of the present invention will be described in detail below with reference to the drawings. The above-mentioned FIG. 1 is a principle configuration diagram of the present invention, and FIG.
FIG. 13 is a diagram showing an embodiment of the present invention, showing an example of an expression form of each event in a search space including a time axis, and FIG. 2 is an example of an expression in the case of a constant event. FIG. 3 shows an example of expression in the case of a time-varying event, FIG. 4 shows an example of expression in the case of moving means for scheduled operation, and FIG. 5 shows a case of constant-speed moving means that can be used at any time. Figure 6 shows an example of expression
Shows an example of expression when there are many constant speed moving means available at an arbitrary time, FIG. 7 shows an example of expression of constant speed moving means called and used at an arbitrary time, and FIG. 8 shows accurate prediction. FIG. 9 shows an example of representation of a possible moving object (moving object with constant velocity motion), FIG. 9 shows an example of representation of a moving object (constant velocity motion) that can be predicted stochastically, and FIG. 11 shows an example of an expression when a moving object is branched, FIG. 11 shows an example of an expression when a branch of a moving object that can be predicted stochastically is replaced with a probability distribution that is temporally fixed, and FIG. FIG. 13 shows an example of an expression when a moving object that can be predicted in advance is detected by sensing, and FIG. 13 shows an example of an expression when the traveling route of a mobile robot is sandwiched between moving objects that can be predicted stochastically. There is.

In the present invention, a map and information of a moving object are described, and several route candidates are found in a two-dimensional or three-dimensional search space used for route search of a mobile robot. A route search method for determining a route with the minimum cost and searching for a route along which the mobile robot moves, by adding a time axis to the two-dimensional or three-dimensional search space axis, A means for processing a time-varying event by replacing it with an event on the spatial axis, specifically,
In a search space including a time axis, an event in which the location does not change with time, an event that occurs accurately at a fixed time, a transportation means that is regularly operated in accordance with a timetable such as a train, is displayed in the time axis. In the including search space, by pseudo-simulating to other events that can be expressed in the search space, the pseudo-event is also taken into consideration, the cost of the lattice of the search space or the cubic lattice is evaluated, and the route search is performed. The means for carrying out are those necessary for carrying out the present invention. The same reference numerals indicate the same objects throughout the drawings.

2 to 13 with reference to FIG.
The route search method in which the time axis of the present invention is also incorporated in the search space, and the configuration and operation of the apparatus will be described. Figure 1
Shows an example of the configuration of a route search device incorporating the time axis in the search space according to the present invention.

That is, the route searching device of the present invention is provided with a map,
In the two-dimensional or three-dimensional search space used to search the path of the mobile robot, which describes the information of the moving object, finds some candidates for the path and determines the path with the lowest cost. , A route search device for searching a route on which the mobile robot moves, a database 1 storing a map, a timetable of the moving object, a law of motion of the moving object, and a probability distribution when the mobile robot moves. A simulation unit 2 that generates a map made up of a grid or a cubic grid with prediction values, and a grid made up of a probability distribution prediction value for the mobile robot, or a map consisting of a cubic grid, a cost grid, Alternatively, the cost calculation unit 3 that converts the map to a cubic grid and the search space that includes the above time axis finds several route candidates and determines the travel route with the lowest cost. Composed of the route search unit 4 for. (Corresponding to the invention described in claim 16) Hereinafter, an example of processing of each event in the search space incorporating the time axis will be described using the route search device. Here, for convenience of description, the space axis is represented by x, or x, y, and the time axis is represented by t.

Constant event: (corresponding to the invention according to claims 1 and 2) The constant event here is the existence of an object (obstacle) whose existence position does not change with time. In the present invention, if such an object is captured in a three-dimensional space including the time axis shown in FIG. 2, it can be treated as a wall or a pillar as shown in the figure, and the route search can be performed. It can be carried out. Therefore, in the search space shown,
The mobile robot can move between the slits to move along the path with the lowest cost.

Time-varying event: (corresponding to the invention described in claims 1 and 3) The time-varying event referred to here is an event that is known to occur accurately at a fixed time. Examples include entrances and exits that are opened and closed at regular times every day, and intersections with traffic lights. FIG. 3 shows an example of a doorway that opens and closes at a fixed time. If such a doorway is regarded as a three-dimensional space including the time axis t, it can be treated as a window, and a route search can be performed. You can Therefore, in the illustrated search space, the mobile robot can move the path with the lowest cost through its window.

Means of regular operation: (corresponding to the inventions of claims 1 and 4) This is an example of a means of regular operation in accordance with a timetable such as a train. FIG. 4 shows an example of representation of such moving means in a three-dimensional space including the time axis t. In this way, such transportation means
When viewed as the three-dimensional space, as shown in the figure, it is formed when a tunnel, that is, after waiting a predetermined time after getting on the moving means, moving at a predetermined speed and then stopping for a predetermined time. The mobile robot can move along a path with the lowest cost through the tunnel.

Constant speed moving means available at any time:
(Corresponding to the invention of claims 1 and 5) This moving means,
It is a moving means that moves at a constant speed that can be used at any time, such as a belt conveyor or a moving sidewalk. When this moving means is regarded as a three-dimensional space including the time axis t, as shown in FIG. 5, it can be handled as a corridor or a tunnel with a restricted moving direction, and the moving robot can be handled. Can be moved using the moving means, and the locus of the moving route at that time is, for example, as shown in FIG.

Approximately, the above-mentioned moving means is regarded as one in which the moving means operating on a regular basis operates continuously,
When viewed as a three-dimensional space including the time axis t, FIG.
As shown in, it is possible to emulate as having many tunnels in close contact with each other. (Corresponding to the invention described in claims 1 and 6) A transportation means that is called at an arbitrary time and used:
Corresponding to the invention described in 7) This moving means is a moving means that is called at an arbitrary time and used, such as an elevator. In this case, probabilistic treatment is necessary because the time from calling to availability is uncertain. Since it is used at a constant speed after being called, as shown in FIG.
It can be used at any time, and can be handled in the same way as a means of moving at a constant speed. The actual movement trajectory of the mobile robot is as shown by the solid line in FIG. That is, the part using the moving means is
It is divided into a part that is waiting by calling the moving means and a part that is moving at a constant speed by using the called moving means.

From the viewpoint of route planning, once it has been decided to use the moving means, it is not necessary to consider the details of its use assumption, and the moving distance and the movable distance using the moving means, Only the required time (corresponding to the cost) at that time becomes a problem. Therefore, as shown in FIG. 7, these movement processes can be simulated by using an apparent speed, similar to a moving means that operates at a constant speed that can be used at any time. it can. However, since the waiting time portion is uncertain, it is necessary to stochastically handle the apparent moving speed indicated by the dotted line by calculating the average waiting time.

Accurately predictable moving object: (claim 1,
(Corresponding to the invention described in 8) This moving object is moving according to a timetable or a strict moving rule, and the movement that can be accurately predicted after the sensing is detected. It is an object (obstacle). When such a moving object is regarded as a three-dimensional space including the time axis t, it can be simulated as a columnar obstacle as shown in FIG.

Probabilistically predictable moving object: (corresponding to the invention according to claims 1 and 9) Such a moving object cannot be predicted accurately, but the database shown in FIG. It is a moving object that can be stochastically predicted by the simulation in the simulation unit 2 based on 1 and information on the moving object obtained by sensing. (See FIGS. 9 to 12) First, an example will be given for a moving object that moves at a constant speed.
When a moving object (obstacle) moving at a constant velocity is captured in a three-dimensional space including the time axis t, the moving object (obstacle) is composed of a cone or a combination thereof as shown in FIGS. 9 to 12. It can be simulated as an obstacle. In FIGS. 9 to 12, the boundary line is depicted as a clear cone, but in reality, the boundary line as described above is determined depending on the prediction accuracy of the moving object and the risk of contact with the moving object. It is necessary to properly use the method of simulating as a clear cone of and the method of simulating as a cone having an unclear boundary line distribution.

FIG. 9 shows an example of prediction of a moving object (obstacle) which moves at a constant speed. What is simulated by the cone is
This is because the accuracy of the sensed speed is poor and there is uncertainty. (Corresponding to the invention described in claims 9, 10 and 11) Fig. 10 is an example of prediction in the case where a moving object moving at a constant speed may branch at a certain time. In an actual application field, it does not move freely in a plane, and when there is an obstacle in front of the moving object (for example, a person, a train, etc.) in the traveling direction, or when an intersection of a corridor is reached, The prediction branch of the traveling direction occurs. Prediction of movement after the occurrence of the branch is based on the total probability of the stochastic movement prediction in each branching direction and the stochastic prediction of which direction the branching will take. ), The probability distribution of the moving position is calculated. (Corresponding to the invention described in claims 9 and 12) FIG. 11 is an example of prediction in the case where there are many possibilities of branching of a moving object that moves at a constant speed. It is not realistic to perform simulation for all the possibilities of all branches because the amount of calculation becomes enormous and the value of the probability of each branch decreases. , When the threshold value is less than or equal to the set threshold value, it is replaced with a probability distribution that is fixed in time as shown in FIG. 11 (specifically, the fixed probability that the mobile robot can move with respect to the above-mentioned lattice and cubic lattice). Must be set). (Corresponding to the invention described in claims 9 and 13) FIG. 12 shows that when a branch of a moving object moving at a constant speed is predicted, the predicted moving object is detected by sensing at a certain time. However, this is an example of the case where the position is confirmed. In this case, when the position is determined, the subsequent prediction of the moving object whose position has been predicted until then is stopped, and a prediction simulation of the detected moving object, that is, , The value of the probability that the mobile robot can move with respect to the above-mentioned lattice and cubic lattice is started. (Corresponding to the invention described in claims 9 and 14) FIG. 13 shows a case of moving to a moving object (obstacle) in which the moving position can be predicted stochastically in an environment where the moving direction is restricted such as a road or a corridor. The state where the movement path of the robot is sandwiched is shown. In this case, the obstacle can be simulated as a tunnel that narrows toward the beginning, and the route search of the mobile robot can be performed. (Corresponding to the invention described in claims 9 and 15) As described above, the route search method and device in which the time axis of the present invention is also incorporated in the search space, the map and the information of the moving object are described. Used for robot route search 2
A time axis t is added to a three-dimensional or three-dimensional search space axis, and a time-varying event is replaced with an event on the space axis for processing. An object whose existence position does not change with time in a two-dimensional space is simulated as a wall or a pillar in a three-dimensional space including the time axis t. Further, an event that occurs exactly at a fixed time, for example, an entrance / exit and an intersection with a traffic light is simulated as a window in a three-dimensional space including the time axis t.
In addition, like a train, a moving means that is operated on a regular basis according to a timetable is simulated as a tunnel in a three-dimensional space including the time axis t. In addition, like a moving walkway, a moving means that moves at a constant speed that can be used at any time is simulated in a three-dimensional space including the time axis t into a corridor or a tunnel with a restricted moving direction, The feature is that the route of the mobile robot is searched.

[0034]

As described above in detail, according to the present invention, the time axis is incorporated in the search space for the route search in the environment including the time-varying event, and By handling it by replacing it with an event, the event that changes with time is simulated as an object that does not change with time. S A route search method for an environment that does not change with time, for example, a known maze method is used. Can be used.
In addition, there is an effect that avoiding a collision with a moving object (obstacle) or the like in the environment and the route search can be unified to perform the route search.

[Brief description of drawings]

FIG. 1 is a block diagram of the principle of the present invention.

FIG. 2 is a diagram showing an embodiment of the present invention (No. 1).

FIG. 3 is a diagram showing an embodiment of the present invention (part 2).

FIG. 4 is a diagram showing an embodiment of the present invention (part 3).

FIG. 5 is a diagram showing an embodiment of the present invention (No. 4).

FIG. 6 is a diagram showing an embodiment of the present invention (No. 5).

FIG. 7 is a diagram showing an embodiment of the present invention (No. 6).

FIG. 8 is a diagram showing an embodiment of the present invention (No. 7).

FIG. 9 is a view showing an embodiment of the present invention (No. 8).

FIG. 10 is a view showing an embodiment of the present invention (No. 9).

FIG. 11 is a diagram showing an embodiment of the present invention (No. 10).

FIG. 12 is a diagram showing an example of the present invention (Part 11).

FIG. 13 is a diagram showing an example of the present invention (part 12).

FIG. 14 is a diagram showing a configuration example of a conventional mobile robot.

[Explanation of symbols]

1 Database 2 Simulation part 3 Cost calculation part 4 Route search part

Claims (16)

[Claims] Claims: 1. A map or information of a moving object is described, and some route candidates are found in a two-dimensional or three-dimensional search space used for a route search of a mobile robot, and the cost is calculated. Is a route search method for determining a minimum route of the mobile robot and searching a route traveled by the mobile robot, wherein a time axis is added to the two-dimensional or three-dimensional search space axis to temporally A route search method that incorporates a time axis into a search space, which is characterized in that changing events are processed by replacing them with events on the space axis. 2. A route search method incorporating the time axis according to claim 1 in a search space, wherein a constant event whose existence position does not change with time is simulated in a fixed object in a search space including the time axis. A route search method that incorporates a time axis into the search space, characterized by performing a route search. 3. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein a time-varying event that accurately occurs at a fixed time is simulated in a window in a search space including the time axis. A route search method that incorporates a time axis into the search space, characterized by performing a route search. 4. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein the means for performing scheduled operation comprises:
In a search space including a time axis, a route search method incorporating a time axis into the search space, which is characterized by simulating a tunnel and performing a route search. 5. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein the constant velocity moving means used at an arbitrary time is restricted in the moving direction in the search space including the time axis. Simulate a corridor or a tunnel
A route search method that incorporates a time axis into the search space, which is characterized by performing a route search. 6. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein a constant speed moving means used at an arbitrary time is approximately a tunnel in a search space including the time axis. A route search method that incorporates a time axis into the search space, which is characterized by performing a route search by imitating a large number of objects that are in close contact with each other. 7. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein a moving means that is called and used at an arbitrary time is probabilistically apparent in a search space including the time axis. A route search method that incorporates a time axis into the search space, which is characterized by performing a route search by simulating a constant speed moving means that has the speed of, and is used at an arbitrary time. 8. A route search method incorporating the time axis according to claim 1 in a search space, wherein a moving object that can be accurately predicted is simulated as a columnar obstacle in a search space including the time axis. , A route search method that incorporates a time axis into the search space, which is characterized by performing a route search. 9. A route search method in which the time axis according to claim 1 is incorporated in a search space, wherein a moving object that can be predicted stochastically in a search space including the time axis is a cone or a combination thereof. A route search method that incorporates a time axis into the search space, characterized by performing a route search by simulating a constructed obstacle. 10. A route search method in which the time axis according to claim 9 is incorporated in a search space, wherein a moving object that can be predicted stochastically, in the search space including the time axis, the cone or a combination thereof. When imitating an obstacle composed of
A route search method that incorporates a time axis into the search space, which is characterized by performing a route search using a clear boundary cone. 11. A route search method in which the time axis according to claim 9 is incorporated in a search space, wherein a moving object that can be predicted stochastically, in the search space including the time axis, the cone or a combination thereof. When imitating an obstacle composed of
A route search method in which a time axis is incorporated into a search space, which is characterized by performing a route search by simulating a conical shape with an unclear boundary line using the existence probability distribution with the estimated traveling direction as the central axis. 12. A route search method in which the time axis according to claim 9 is incorporated in a search space, wherein a moving object that can be predicted stochastically, in the search space including the time axis, the cone or a combination thereof. When imitating an obstacle composed of
When a plurality of traveling directions exist in the traveling direction, a time axis characterized by performing a route search by imitating an obstacle that branches at a predetermined probability with respect to the plurality of traveling directions is incorporated into the search space. Route search method. 13. A route search method in which the time axis according to claim 12 is incorporated in a search space, and when the number of branches is large, the route is simulated by an obstacle having a temporally fixed probability distribution. A route search method that incorporates a time axis into the search space characterized by conducting a search. 14. A route search method in which the time axis according to claim 9 is incorporated in a search space, wherein a moving object that can be predicted stochastically, in the search space including the time axis, the cone or a combination thereof. When imitating an obstacle composed of
When the predicted moving object is detected by sensing at a certain time, the obstacle simulated before is eliminated at the time when the position is determined, and a new cone based on the probability distribution is created. A route search method that incorporates a time axis into the search space, which is characterized by performing a route search by simulating an obstacle composed of. 15. A route search method in which the time axis according to claim 9 is incorporated in a search space, and the route is sandwiched between moving objects that can be predicted stochastically in an environment where the moving direction is restricted such as roads and corridors. When a route search is performed in a closed state, the moving object is simulated in a search space including the time axis into a tunnel that narrows toward the beginning, and the route search is performed. The route search method introduced in. 16. A method for finding the cost of several routes in a two-dimensional or three-dimensional search space used for a route search of a mobile robot in which a map and information of a moving object are described. Is a route search device for determining a minimum route of a mobile robot and searching for a route along which the mobile robot moves. A database unit (1) storing a map, a timetable of a moving object, a law of motion of a moving object, etc. And a simulation unit (2) that generates a map composed of a grid or a cubic grid with a probability distribution prediction value for a mobile robot,
A cost calculation unit (3) for converting a map composed of a grid or a cubic grid with a probability distribution prediction value for the mobile robot into a cost grid or a map composed of a cubic grid.
And a route search unit (4) that finds several route candidates in the search space including the above time axis and determines the travel route with the lowest cost. Incorporated route search device.