JP2021096703A - Travel control device - Google Patents

Abstract

To provide a travel control device that is able to improve estimation accuracy of a self-position of a moving body.SOLUTION: In at least one of the detection range of a laser sensor 3A and the detection range of a laser sensor 3B, there may be various causes of decreasing estimation accuracy of a self-position. To counteract this, a position estimation unit 11 acquires information about a decrease in estimation accuracy of a self-position and, based on the information, adjusts a use mode for a detection result from the laser sensor 3A and a use mode for a detection result from the laser sensor 3B. Thus, the position estimation unit 11 can adjust the use mode for the detection result from the laser sensor 3A and the detection result from the laser sensor 3B so as to inhibit a decrease in estimation accuracy of the self-position.SELECTED DRAWING: Figure 1

Description

The present invention relates to a traveling control device.

As a conventional traveling control device, for example, a technique described in Patent Document 1 is known. The travel control device described in Patent Document 1 detects a surrounding object and travels the moving body while estimating the self-position of the moving body based on the detection result. This travel control device estimates the self-position of the moving body by detecting surrounding objects with a detection unit such as a laser sensor.

Japanese Unexamined Patent Publication No. 2011-253414

By the way, as in the above-mentioned prior art, in a traveling control device that detects a surrounding object and travels the moving object while estimating the self-position of the moving object based on the detection result, it is registered in the detected object and map data. Compare with the object that was made. However, there are various causes around the moving body that reduce the accuracy of estimating the self-position. As an example, in addition to an object registered in the map data, there may be an object not registered in the map data such as a person or another moving object around the moving object. In this case, the travel control device erroneously recognizes a person, a moving body, or the like as an object registered in the map data. As a result, the accuracy of estimating the self-position of the moving body of the traveling control device may decrease.

An object of the present invention is to provide a traveling control device capable of improving the estimation accuracy of the self-position of a moving body.

One aspect of the present invention is a travel control device for traveling a moving body, in which a first detection unit and a second detection unit for detecting an object around the moving body, and a first detection unit and a second detection unit are used. Based on the detection result by at least one of the above and the map data acquired in advance, the position estimation unit that estimates the self-position of the moving body and the self-position of the moving body estimated by the position estimation unit. A traveling control unit for controlling traveling is provided, and the position estimation unit acquires information on a decrease in the estimation accuracy of the self-position, and based on the information, the first detection result by the first detection unit and the second detection unit. The usage mode of the second detection result by the detection unit of is adjusted.

In such a traveling control device, the position estimation unit estimates the self-position of the moving body based on the detection result by at least one of the first detection unit and the second detection unit and the map data acquired in advance. To do. As a result, the position estimation unit can estimate the self-position of the moving body by comparing the object detected by the detection unit with the object registered in the map data. Here, in at least one of the detection range of the first detection unit and the detection range of the second detection unit, there may be various causes that reduce the estimation accuracy of the self-position. On the other hand, the position estimation unit acquires information on the decrease in the estimation accuracy of the self-position, and based on the information, the first detection result by the first detection unit and the second detection by the second detection unit. Adjust the usage of the results. As a result, the position estimation unit adjusts the usage mode of the first detection result by the first detection unit and the second detection result by the second detection unit so as to suppress the decrease in the estimation accuracy of the self-position. can do. From the above, the travel control device can improve the estimation accuracy of the self-position of the moving body.

When the position estimation unit estimates its own position using one of the first detection result and the second detection result, and acquires information on a decrease in estimation accuracy for the other detection result. , The self-position may be estimated by switching to the other detection result. As a result, since only one of the first detection unit and the second detection unit is used while suppressing a decrease in the estimation accuracy of the self-position, the processing load can be reduced.

The position estimation unit may acquire a determination result indicating that the estimation accuracy has decreased as information regarding the decrease in the estimation accuracy. As a result, it is possible to adjust the usage mode of the detection unit based on the fact that the estimation accuracy is actually lowered.

The position estimation unit may acquire information indicating that there is an obstacle that reduces the estimation accuracy around the moving body as information regarding the decrease in the estimation accuracy. The position estimation unit can quickly adjust the usage mode of the detection unit based on the presence of an obstacle without calculating that the estimation accuracy has actually decreased.

The position estimation unit may acquire information indicating that the moving body has passed the position where the estimation accuracy is lowered as the information regarding the deterioration of the estimation accuracy. The position estimation unit can quickly adjust the usage mode of the detection unit based on the passing position of the moving body without calculating that the estimation accuracy is actually lowered.

According to the present invention, the accuracy of estimating the self-position of the moving body can be improved.

It is a schematic block diagram which shows the traveling control apparatus which concerns on one Embodiment of this invention. It is a conceptual diagram which shows how a moving body runs in a warehouse or a factory. It is a conceptual diagram which shows the state when a position estimation part estimates a self-position based on a detection result. It is a flowchart which shows the content of self-position estimation processing by a position estimation part. It is a conceptual diagram which shows the state when the position estimation part of the traveling control device which concerns on a modification, estimates self-position based on a detection result. It is a flowchart which shows the content of the self-position estimation process by the position estimation part of the traveling control device which concerns on a modification. It is a conceptual diagram which shows the state when the position estimation part of the traveling control device which concerns on a modification, estimates self-position based on a detection result. It is a flowchart which shows the content of the self-position estimation process by the position estimation part of the traveling control device which concerns on a modification. It is a conceptual diagram which shows the state when the position estimation part of the traveling control device which concerns on a modification, estimates self-position based on a detection result. It is a flowchart which shows the content of the self-position estimation process by the position estimation part of the traveling control device which concerns on a modification.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing a traveling control device 1 according to an embodiment of the present invention. In FIG. 1, the travel control device 1 of the present embodiment is a device that automatically travels the moving body 2 as shown in FIG. Note that FIG. 2 is a conceptual diagram showing how the moving body 2 travels in a structure such as a warehouse or a factory. The moving body 2 is, for example, a forklift, an automatic guided vehicle, or the like. The travel control device 1 is mounted on the moving body 2.

The travel control device 1 includes a laser sensor 3A (first detection unit), a laser sensor 3B (second detection unit), a drive unit 4, and a controller 5.

The laser sensors 3A and 3B detect an object around the moving body 2. Further, the laser sensors 3A and 3B measure the positional relationship with the detected object. The laser sensors 3A and 3B irradiate the periphery of the moving body 2 with a laser and receive the reflected light of the laser to detect an object around the moving body 2 from the laser sensors 3A and 3B, and the distance to the object. To measure. As the laser sensors 3A and 3B, for example, a laser range finder is used. The laser to be used may be a 2D laser or a 3D laser.

The laser sensors 3A and 3B are provided so as to face different directions from each other. As a result, the detection range of the laser sensor 3A and the detection range of the laser sensor 3B are set to different ranges. For example, as shown in FIG. 2, the laser sensor 3A is provided on the front side of the moving body 2. Further, the laser sensor 3A irradiates the laser L in a fan shape toward the front of the moving body 2. Specifically, the laser sensor 3A irradiates the laser L in a predetermined angle range centered on the forward straight direction of the moving body 2. As a result, the laser sensor 3A can perform detection with a predetermined angle range in front of the moving body 2 as a detection range. The laser sensor 3B is provided on the rear side of the moving body 2. The laser sensor 3B irradiates the laser L in a fan shape toward the rear of the moving body 2. Specifically, the laser sensor 3B irradiates the laser L in a predetermined angle range centered on the backward straight direction of the moving body 2. As a result, the laser sensor 3B can perform detection with a predetermined angle range behind the moving body 2 as a detection range.

The laser L emitted from the laser sensors 3A and 3B hits the object 20, and the laser (reflected light) reflected by the object 20 is received by the laser sensors 3A and 3B. The object 20 includes a feature 21 which is a characteristic structure suggesting a self-position, such as a stationary object such as a wall 26 or a pillar 27. Such a feature 21 is registered in the map data. Further, the object 20 includes a dynamic object 22 such as a person such as a worker or another moving object (see FIG. 3). Such a dynamic object 22 is not registered in the map data and becomes an obstacle in estimating the self-position. In addition to dynamic objects, stationary objects such as temporarily placed luggage and temporary objects may also be obstacles.

Although not particularly shown, the drive unit 4 includes a traveling motor that rotates the traveling wheels of the moving body 2 and a steering motor that steers the steering wheels of the moving body 2.

The controller 5 is composed of a CPU, RAM, ROM, an input / output interface, and the like. The controller 5 includes a position estimation unit 11, a travel control unit 12, and a storage unit 13.

The position estimation unit 11 estimates the self-position of the moving body 2 based on the detection results of the laser sensors 3A and 3B and the map data acquired in advance. The position estimation unit 11 estimates the self-position of the moving body 2 by using, for example, a SLAM (simultaneous localization and mapping) method. SLAM is a self-position estimation technology that performs self-position estimation using sensor data and map data. SLAM simultaneously estimates the self-position and creates an environmental map. Specifically, the position estimation unit 11 matches the distance data to the object 20 detected by the laser sensor 3 with the map data of the surrounding environment of the moving body 2 to estimate the self-position of the moving body 2. Do.

The position estimation unit 11 acquires information on the decrease in the estimation accuracy of the self-position, and based on the information, the detection result by the laser sensor 3A (first detection result) and the detection result by the laser sensor 3B (second detection result). The usage mode of the detection result) is adjusted. The information regarding the decrease in the estimation accuracy of the self-position includes information indicating that the estimation accuracy of the self-position has decreased and information indicating items that may cause the estimation accuracy of the self-position to decrease. The information indicating that the estimation accuracy of the self-position has decreased is information that can quantitatively indicate the decrease in the estimation accuracy, such as an index indicating the degree of matching between the detection result and the map data. Information indicating items that may cause a decrease in the estimation accuracy of the self-position includes information indicating the existence of an obstacle that causes a decrease in the estimation accuracy around the moving body 2 and estimation around the moving body 2. This is information indicating that a state that causes a decrease in accuracy has occurred. Obstacles include dynamic objects such as people and other moving objects, and static objects such as luggage and temporary objects. Examples of the state that causes a decrease in the estimation accuracy include a change in the position of the feature 21. In the present embodiment, the position estimation unit 11 has changed the position of the information indicating that there is an obstacle that reduces the estimation accuracy around the moving body 2 and the feature 21 as the information regarding the decrease in the estimation accuracy. Get the information that indicates.

The usage mode of the detection result by the laser sensor 3A (first detection result) and the detection result by the laser sensor 3B (second detection result) is adjusted. Adjusting the usage mode includes switching from a state in which one laser sensor is used to estimate the self-position to the other laser sensor to estimate the self-position. Further, in order to adjust the usage mode, in a state where both laser sensors 3A and 3B are performing detection, one of the detection results is adopted to estimate the self-position, and the other detection result is not used. included. In addition, adjusting the usage mode also includes estimating the self-position using both laser sensors from the state where the self-position is estimated using one of the laser sensors. In the present embodiment, the position estimation unit 11 estimates the self-position using one of the detection results of the laser sensor 3A and the detection result of the laser sensor 3B, and estimates for the one detection result. When information on the decrease in accuracy is acquired, the self-position is estimated by switching to the other detection result.

For example, in the example shown in FIG. 3, the objects 20A to 20D exist in front of the moving body 2. Of these, the objects 20A, 20C, and 20D are the feature objects 21, and the object 20B is the dynamic object 22. The position estimation unit 11 identifies the object 20B, which is the dynamic object 22, as an obstacle among the objects 20A to 20D. Further, the object 20C, which is the feature 21, identifies that the position has been changed from the map data. From this, the position estimation unit 11 acquires information indicating that an obstacle that lowers the estimation accuracy exists in front of the moving body 2 and that the feature 21 has been repositioned. The position estimation unit 11 switches from the laser sensor 3A on the front side to the laser sensor 3B on the rear side to estimate its own position. Specifically, the position estimation unit 11 includes an information acquisition unit 14, an estimation accuracy confirmation unit 16, and a calculation unit 17.

The information acquisition unit 14 receives various information necessary for estimating its own position. The information acquisition unit 14 acquires the detection results of the laser sensors 3A and 3B. When the laser sensor 3A on the front side is activated, the information acquisition unit 14 acquires the detection result by the laser sensor 3A. When the laser sensor 3B on the rear side is activated, the information acquisition unit 14 acquires the detection result by the laser sensor 3B. As shown in FIG. 3A, when the laser L irradiates the objects 20A to 20D, the laser sensor 3A acquires the position information of the point clouds PA to PD. As a result, the laser sensor 3A detects the objects 20A to 20D based on the position information of the point clouds PA to PD, and acquires the positional relationship of the objects 20A to 20D with the moving body 2. The information acquisition unit 14 acquires the detection results of these laser sensors 3. Note that FIG. 3 shows only the laser L that irradiates the edges and the like of each object.

The estimation accuracy confirmation unit 16 confirms the estimation accuracy when the self-position is estimated based on the detection results of the laser sensors 3A and 3B. The estimation accuracy confirmation unit 16 acquires information regarding a decrease in the estimation accuracy of the self-position.

Specifically, the estimation accuracy confirmation unit 16 identifies an obstacle from the objects detected by the laser sensor 3A. For example, the estimation accuracy confirmation unit 16 can grasp whether or not the moving directions of the point clouds PA to PD are changing by grasping the positions of the point clouds PA to PD in time series. Since the feature 21 is stationary, the point clouds PA, PC, and PD corresponding to the feature 21 move relative to each other in the moving direction according to the movement of the moving body 2. Therefore, the estimation accuracy confirmation unit 16 identifies the objects 20A, 20C, and 20D corresponding to the point clouds PA, PC, and PD as the feature 21 required for estimating the self-position. On the other hand, since the dynamic object 22 is moving, the point cloud PB corresponding to the dynamic object 22 moves relatively in a direction different from the movement of the moving body 2. Therefore, the estimation accuracy confirmation unit 16 identifies the object 20B corresponding to the point cloud PB as the dynamic object 22 which is an obstacle. In this way, the estimation accuracy confirmation unit 16 acquires the information that an obstacle exists in front of the estimation accuracy as information that the estimation accuracy of the self-position may decrease.

Further, the estimation accuracy confirmation unit 16 specifies that the object 20C has moved from the position registered in the map data among the objects 20A, 20C, and 20D as the feature 21 detected by the laser sensor 3A. The estimation accuracy confirmation unit 16 compares the position of the object 20C determined from the point cloud PC with the position of the object 20C in the map data (the position indicated by the virtual line in FIG. 3A), thereby causing the object 20C to move. Identify that you are moving from the position at the time of registration. In this way, the estimation accuracy confirmation unit 16 acquires the information that the feature 21 moved from the position registered in the map data is present ahead as information that the estimation accuracy of the self-position can be lowered.

The calculation unit 17 performs a calculation for estimating the self-position of the moving body 2 from the detection results of the laser sensors 3A and 3B. Further, the calculation unit 17 adjusts the usage mode of the detection result by the laser sensor 3A and the detection result by the laser sensor 3B based on the information regarding the decrease in the estimation accuracy of the self-position. When the calculation unit 17 acquires information on the decrease in the estimation accuracy of the self-position while using the front laser sensor 3A, the calculation unit 17 switches to the rear laser sensor 3B to estimate the self-position. When switching from the front laser sensor 3A to the rear laser sensor 3B, the calculation unit 17 uses the point clouds PE and PF corresponding to the rear objects 20E and 20F as shown in FIG. 3 (b). To estimate the self-position. When the estimation accuracy confirmation unit 16 acquires information on the decrease in the estimation accuracy of the self-position while using the rear laser sensor 3B, the estimation accuracy confirmation unit 16 switches to the front laser sensor 3A to estimate the self-position. ..

Specifically, the calculation unit 17 performs an estimation calculation of the self-position of the moving body 2 using the point clouds PE and PF and the map data stored in the storage unit 13. The self-position estimation by the calculation unit 17 may be performed by using the SLAM method. Further, the calculation unit 17 may detect the objects 20E and 20F from the point clouds PE and PF and estimate the self-position by matching the distance data between the object and the moving body and the map data of the environment around the moving body. ..

The travel control unit 12 controls the travel of the moving body 2 based on the self-position of the moving body 2 estimated by the position estimation unit 11. The travel control unit 12 calculates the movement direction and the movement distance of the moving body 2 by comparing the self-position with the target position. Further, the travel control unit 12 outputs a control signal to the drive unit 4 so that the moving body 2 heads for the target position based on the calculation result. The storage unit 13 stores various data. The storage unit 13 stores map data.

Next, with reference to FIG. 4, the self-position estimation process of the moving body 2 by the position estimation unit 11 will be described. FIG. 4 is a flowchart showing the contents of the self-position estimation process by the position estimation unit 11. Here, it is assumed that the process is started from the state where the self-position is estimated by the laser sensor 3A on the front side using the detection result in front of the moving body 2. First, the information acquisition unit 14 of the position estimation unit 11 acquires the detection result by the laser sensor 3A (step S100). In the example shown in FIG. 3A, the information acquisition unit 14 acquires the detection results of the objects 20A to 20D detected based on the point clouds PA to PD.

Next, the estimation accuracy confirmation unit 16 of the position estimation unit 11 identifies an obstacle from the objects detected by the laser sensor 3 (step S110). In the example shown in FIG. 3A, the estimation accuracy confirmation unit 16 identifies the object 20B as an obstacle. Next, the estimation accuracy confirmation unit 16 determines whether or not an obstacle exists in the detection range of the laser sensor 3 (step S120).

When it is determined in step S120 that an obstacle exists, the calculation unit 17 switches from the laser sensor 3A on the front side to the laser sensor 3B on the rear side, and based on the detection result by the laser sensor 3B, the self of the moving body 2. The position is estimated (step S130). In the example shown in FIG. 3, the calculation unit 17 estimates the self-position of the moving body 2 by using the point clouds PE and PF corresponding to the rear objects 20E and 20F. When the process of step SS130 is completed, the process shown in FIG. 4 is completed. If an obstacle is identified while the rear laser sensor 3B is estimating the self-position, the calculation unit 17 switches from the rear laser sensor 3B to the front laser sensor 3A and switches the laser. The self-position of the moving body 2 is estimated based on the detection result by the sensor 3A.

When it is determined in step S120 that the obstacle does not exist in front, the calculation unit 17 estimates the self-position of the moving body 2 based on the detection result by the laser sensor 3A without switching (step S140). When the process of step S140 is completed, the process shown in FIG. 4 is completed.

Next, the operation / effect of the traveling control device 1 according to the present embodiment will be described.

First, a traveling control device according to a comparative example will be described. The travel control device according to the comparative example detects an object in front and travels the moving body while estimating the self-position of the moving body based on the detection result. At this time, the travel control device compares all the objects detected ahead with the objects registered in the map data. However, in addition to the objects registered in the map data, there may be objects such as people and other moving objects that are not registered in the map data around the moving body. In this case, the travel control device according to the comparative example erroneously recognizes a person, a moving body, or the like as an object registered in the map data. For example, when the traveling control device mistakenly recognizes another moving body as a wall, the distance between the self-position and the other moving body is erroneously estimated as the distance between the self-position and the wall. As a result, the accuracy of estimating the self-position of the moving body of the traveling control device may decrease.

On the other hand, in the traveling control device 1 according to the present embodiment, the position estimation unit 11 self of the moving body 2 based on the detection result by at least one of the laser sensors 3A and 3B and the map data acquired in advance. Estimate the position. As a result, the position estimation unit 11 can estimate the self-position of the moving body 2 by comparing the object detected by the laser sensors 3A and 3B with the object registered in the map data. Here, in at least one of the detection range of the laser sensor 3A and the detection range of the laser sensor 3B, there may be various causes that reduce the estimation accuracy of the self-position. On the other hand, the position estimation unit 11 acquires information regarding the decrease in the estimation accuracy of the self-position, and adjusts the usage mode of the detection result by the laser sensor 3A and the detection result by the laser sensor 3B based on the information. As a result, the position estimation unit 11 can adjust the usage mode of the detection result by the laser sensor 3A and the detection result by the laser sensor 3B so as to suppress the decrease in the estimation accuracy of the self-position. From the above, the travel control device 1 can improve the estimation accuracy of the self-position of the moving body 2.

The position estimation unit 11 estimates the self-position using one of the detection results of the laser sensor 3A and the detection result of the laser sensor 3B, and provides information on the decrease in estimation accuracy to the one detection result. When it is acquired, it switches to the other detection result and estimates its own position. As a result, since only one of the laser sensor 3A and the laser sensor 3B is used while suppressing a decrease in the estimation accuracy of the self-position, the processing load can be reduced.

The position estimation unit 11 acquires information indicating that there is an obstacle that reduces the estimation accuracy around the moving body 2 as information regarding the decrease in the estimation accuracy. The position estimation unit 11 can quickly adjust the usage mode of the detection unit based on the presence of an obstacle without calculating that the estimation accuracy has actually decreased.

The present invention is not limited to the above-described embodiment.

In the above-described embodiment, information that an obstacle exists, information that the position of the feature 21 has changed, and the like are acquired as information indicating a matter that may cause a decrease in the estimation accuracy of the self-position. Instead of this, information that indirectly indicates the existence of an obstacle may be acquired. When the travel control device 1 detects a dynamic object and tracks the dynamic object, the information that the tracking state is in the tracking state is acquired as information indicating a matter that may cause a decrease in the estimation accuracy of the self-position. Good. For example, as shown in FIG. 5A, when an object 20B as a dynamic object exists in front of the moving body 2, the laser sensor 3A on the front side detects the object 20B as a dynamic object and the object 20B. To track. When the estimation accuracy confirmation unit 16 acquires the information that the laser sensor 3A is in the tracking state, it is assumed that the dynamic object 22 which is an obstacle exists in the detection range of the laser sensor 3A, and the self based on the detection result of the laser sensor 3A. It is judged that the accuracy of position estimation may decrease. In this case, as shown in FIG. 5B, the calculation unit 17 switches from the laser sensor 3A on the front side to the laser sensor 3B on the rear side, and the point clouds PE and PF corresponding to the objects 20E and 20F existing in the rear side. Is used to estimate the self-position.

The self-position estimation process when tracking a dynamic object will be described with reference to FIG. FIG. 6 is a flowchart showing the contents of the self-position estimation process by the position estimation unit 11 of the travel control device according to the modified example. First, the information acquisition unit 14 of the position estimation unit 11 acquires the detection result by the laser sensor 3A (step S100). Further, the estimation accuracy confirmation unit 16 confirms whether or not the laser sensor 3A is in the tracking state (step S210). Next, the estimation accuracy confirmation unit 16 determines whether or not the laser sensor 3A is in the tracking state (step S220). When it is determined in step S220 that the laser sensor 3A is in the tracking state, the calculation unit 17 switches from the front laser sensor 3A to the rear laser sensor 3B and moves based on the detection result by the laser sensor 3B. The self-position of the body 2 is estimated (step S130). When it is determined in step S220 that the laser sensor 3A is not in the tracking state, the calculation unit 17 estimates the self-position of the moving body 2 based on the detection result by the laser sensor 3A without switching (step S140). .. When the process of step S140 is completed, the process shown in FIG. 6 is completed.

In addition, the position estimation unit 11 may acquire a determination result indicating that the estimation accuracy has decreased as information regarding the decrease in the estimation accuracy. As a result, it is possible to adjust the usage mode of the laser sensors 3A and 3B based on the fact that the estimation accuracy is actually lowered. Specifically, as shown in FIG. 7, the travel control device simultaneously performs detection by the front laser sensor 3A and detection by the laser sensor 3B. Then, the position estimation unit 11 confirms the self-position estimation accuracy based on the detection result of the laser sensor 3A and the self-position estimation accuracy based on the detection result of the laser sensor 3B. Then, when the position estimation unit 11 acquires a determination result that the estimation accuracy is lowered by any of the detection results by the laser sensors 3A and 3B, the self-position is estimated using the detection result having the better estimation accuracy. .. The estimation accuracy is determined by using an index or the like indicating the degree of matching between the detection result and the map data.

The self-position estimation process when acquiring the determination result that the accuracy of the estimation result has deteriorated will be described with reference to FIG. FIG. 8 is a flowchart showing the contents of the self-position estimation process by the position estimation unit 11 of the travel control device according to the modified example. First, the information acquisition unit 14 of the position estimation unit 11 acquires the detection results of the laser sensors 3A and 3B, and the calculation unit 17 estimates the self-position using those results (step S300). The estimation accuracy confirmation unit 16 confirms the estimation accuracy based on the detection result of the laser sensor 3A and the estimation accuracy based on the detection result of the laser sensor 3B (step S310). Then, the calculation unit 17 estimates the self-position using the detection result of the laser sensor 3A and the detection result of the laser sensor 3B, whichever has the better self-position estimation accuracy (step S330). When the process of step S330 is completed, the process shown in FIG. 8 is completed.

Further, when the travel control device estimates the self-position using one of the detection results of the laser sensors 3A and 3B and obtains a determination result indicating that the estimation accuracy has deteriorated, the travel control device of the laser sensors 3A and 3B You may switch to the other detection result. For example, as shown in FIG. 9A, the estimation accuracy of the self-position decreases when the same landscape continues. In the vicinity of the entrance of the long corridor, the scenery in front is the same for a while, so it is difficult to estimate how far the moving body 2 is from the entrance only from the detection result of the laser sensor 3A. On the other hand, near the entrance, the rear view is characteristic. Therefore, the position estimation unit 11 estimates its own position near the entrance by using the detection result of the laser sensor 3B behind. Then, in the vicinity of the exit where the feature in the rear landscape disappears, the position estimation unit 11 estimates its own position using the detection result of the laser sensor 3A on the front side. As a result, it is possible to quickly estimate the self-position using the detection result of the laser sensor 3A in front of the exit in the vicinity of the exit where the state in front is characteristic.

The flow of the estimation process in this case will be described with reference to FIG. FIG. 10 is a flowchart showing the contents of the self-position estimation process by the position estimation unit 11 of the travel control device according to the modified example. First, the information acquisition unit 14 of the position estimation unit 11 acquires the detection result of the laser sensor 3A, and the calculation unit 17 estimates its own position using the result (step S400). The estimation accuracy confirmation unit 16 confirms the estimation accuracy based on the detection result of the laser sensor 3A (step S410). Next, the estimation accuracy confirmation unit 16 determines whether or not the estimation accuracy of the self-position based on the detection result of the laser sensor 3A is good (step S420). When it is determined in step S420 that the estimation accuracy is good, the calculation unit 17 estimates the self-position of the moving body 2 based on the detection result by the laser sensor 3A without switching (step S430). If it is determined in step S420 that the estimation accuracy is poor, the front laser sensor 3A is switched to the rear laser sensor 3B, and the self-position of the moving body 2 is estimated based on the detection result by the laser sensor 3B (the self-position of the moving body 2 is estimated). Step S440). When the processes of steps S430 and S440 are completed, the processes shown in FIG. 10 are completed.

The position estimation unit 11 may acquire information indicating that the moving body has passed the position where the estimation accuracy is lowered as the information regarding the deterioration of the estimation accuracy. The position estimation unit 11 can quickly adjust the usage mode of the laser sensors 3A and 3B based on the passing position of the moving body 2 without calculating that the estimation accuracy is actually lowered. As shown in FIG. 9B, the mark 50 is set at a position where the estimation accuracy of the self-position based on the detection result of the rear laser sensor 3B starts to decrease. When the position estimation unit 11 detects the mark 50, it determines that it has passed a position where the estimation accuracy is lowered, and estimates its own position using the detection result of the laser sensor 3A in front. As information that the estimation accuracy of the self-position may decrease, the amount of deviation between the value estimated by the laser detection result and the map and the value for specifying the position using a device other than the laser is equal to or more than a predetermined threshold. In such a case, it may be determined that the estimation accuracy of the self-position is lowered. As a method of specifying a position other than the laser, an AR marker installed on the road surface may be detected by a camera to specify the position.

In the above-described embodiment, the laser sensor has been exemplified as a detection unit for detecting an object around the moving body 2, but the specific configuration of the detection unit is not particularly limited. For example, a camera may be adopted as the detection unit. The camera can detect an object and also detect information such as a distance to the object by performing image processing on the acquired image. For example, an object can be detected by detecting the edge of the object captured in the image. Further, by machine learning the shape of a person and the shape of the machine base in advance, it is possible to identify that the object in the image is a dynamic object such as a person or the machine base. Further, the estimation accuracy confirmation unit 16 may detect a surrounding object by using the image acquired by the camera. Further, the detection unit may be configured by a combination of a laser sensor and a camera. For example, the detection unit may detect the object based on the image of the camera and measure the distance to the object using the result of the laser sensor. In addition, the detection unit may be composed of a sensor such as an ultrasonic wave.

1 ... Travel control device, 2 ... Mobile body, 3 ... Laser sensor (detection unit), 11 ... Position estimation unit, 12 ... Travel control unit, 20 ... Object.

Claims (5)

In a traveling control device that travels a moving body,
A first detection unit and a second detection unit that detect an object around the moving body, and
A position estimation unit that estimates the self-position of the moving body based on the detection results of at least one of the first detection unit and the second detection unit and the map data acquired in advance.
A traveling control unit that controls the traveling of the moving body based on the self-position of the moving body estimated by the position estimating unit is provided.
The position estimation unit
Obtaining information on the decrease in the estimation accuracy of the self-position,
A travel control device that adjusts the usage mode of the first detection result by the first detection unit and the second detection result by the second detection unit based on the information. The position estimation unit
The self-position is estimated by using one of the first detection result and the second detection result.
The travel control device according to claim 1, wherein when the information regarding the decrease in the estimation accuracy is acquired with respect to the one detection result, the self-position is estimated by switching to the other detection result. The travel control device according to claim 1 or 2, wherein the position estimation unit acquires a determination result indicating that the estimation accuracy has decreased as information regarding the decrease in the estimation accuracy. Any one of claims 1 to 3, wherein the position estimation unit acquires information indicating that an obstacle that reduces the estimation accuracy exists around the moving body as information regarding the decrease in the estimation accuracy. The travel control device according to. The position estimation unit according to any one of claims 1 to 4, wherein the position estimation unit acquires information indicating that the moving body has passed the position where the estimation accuracy is lowered as the information regarding the deterioration of the estimation accuracy. Travel control device.

Images