JP7758558B2 - Autonomous construction robot control device, autonomous construction robot control system, and autonomous construction robot control method - Google Patents

Description

The present invention relates to an autonomous construction robot control device, an autonomous construction robot control system, and an autonomous construction robot control method.

Efforts are underway to introduce autonomous robots into construction site management to improve the efficiency of management work. Self-localization is sometimes used to enable the robots to navigate autonomously. Self-localization is important for patrolling construction sites and collecting comprehensive construction data.
Indoor positioning technology is used when estimating self-position in a closed space where radio waves from satellites such as GPS (Global Positioning System) cannot reach. There are various methods for indoor positioning, and the most suitable method varies depending on the purpose, required accuracy, condition of the measurement area, and cost. Recently, efforts to improve the accuracy of indoor positioning by using "hybrid positioning" that combines multiple positioning technologies are becoming mainstream.
In addition, by using SLAM (Simultaneous Localization and Mapping) technology, which simultaneously estimates self-position and creates maps, autonomous operation of unmanned construction machinery inside tunnels is becoming a reality.

When using SLAM technology to operate autonomous robots such as unmanned construction machinery, the autonomous robot needs to understand the situation in the external world. To understand the situation in the external world, sensors such as LiDAR (laser scanners), cameras, and ToF (Time Of Flight) sensors are used. These sensors acquire point cloud data that represents the three-dimensional spatial appearance of the external world.

In addition to these technologies, systems have also been proposed for tracking progress inside buildings under construction (for example, Patent Document 1).

Japanese Patent Application Laid-Open No. 2020-160563

However, when introducing an autonomous robot into an underground space where the inverted construction method is used, for example, it is necessary to grasp the operating status of the construction machinery in addition to its position information in order to avoid contact with other construction machinery operating in the same underground space.With autonomous robots equipped with conventional SLAM technology, there is a concern that they may not be able to adequately predict danger to construction machinery that rotates while remaining stationary (without changing its position information), and may collide (interfere) with the bucket or arm of the construction machinery.
It should be noted that the system of Patent Document 1 allows people to grasp progress information inside a building under construction, but is not intended for autonomous robots to grasp such information.

The present invention was made in light of these circumstances, and its purpose is to provide an autonomous construction robot control device, an autonomous construction robot control system, and an autonomous construction robot control method that enable an autonomous robot to determine whether or not there is an object to be detected in its surroundings.

In order to solve the above-mentioned problems, one aspect of the present invention is an autonomous construction robot control device that includes a sensor data acquisition unit that acquires waveform data from a sensor that is installed in a target area where construction work is to be performed and detects vibrations; an analysis unit that analyzes the frequencies contained in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machinery operating in the target area; a control unit that outputs a control signal to an autonomous construction robot to cause it to operate so as not to interfere with the construction machinery based on the operating status of the construction machinery; and an object position estimation unit that generates position information that estimates the position of the construction machinery based on the analysis results of the analysis unit, wherein the control unit generates a movement path that does not interfere with the estimated position in accordance with the operating status and outputs a control signal representing the generated movement path .

Another aspect of the present invention is an autonomous construction robot control system comprising: a sensor that is installed in a target area where construction work is to be carried out and detects vibrations; a sensor data acquisition unit that acquires waveform data from the sensor; an analysis unit that analyzes the frequencies contained in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machinery operating in the target area; a control unit that outputs a control signal to an autonomous construction robot to cause the robot to operate so as not to interfere with the construction machinery based on the operating status of the construction machinery; and an object position estimation unit that generates position information that estimates the position of the construction machinery based on the analysis results of the analysis unit, wherein the control unit generates a movement path that does not interfere with the estimated position in accordance with the operating status and outputs a control signal representing the generated movement path .

Another aspect of the present invention is a control method for an autonomous construction robot executed by a computer, in which a sensor data acquisition unit acquires waveform data from a sensor that is installed in a target area where construction work is to be performed and detects vibrations, an analysis unit analyzes the frequencies contained in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machinery operating in the target area, a control unit outputs a control signal to the autonomous construction robot based on the operating status of the construction machinery to cause it to operate without interfering with the construction machinery , an object position estimation unit generates position information that estimates the position of the construction machinery based on the analysis results of the analysis unit, and the control unit generates a movement path that does not interfere with the estimated position in accordance with the operating status and outputs a control signal representing the generated movement path .

As described above, this invention allows an autonomous robot to determine whether or not there is a detection target in its surroundings, making it possible to operate the autonomous robot without interfering with the detection target.

1 is a schematic diagram showing the configuration of an autonomous construction robot control system S according to one embodiment. 1 is a schematic functional block diagram showing the configuration of an autonomous construction robot control device 1. FIG. 3 is a diagram showing an example of frequency characteristic data stored in a frequency data storage unit 12. FIG. FIG. 4 is a diagram illustrating an example of waveform data detected by an acceleration sensor. 10 is a flowchart illustrating the operation of the autonomous construction robot control device 1.

An autonomous construction robot control system using an autonomous construction robot control device according to one embodiment of the present invention will be described below with reference to the drawings.
FIG. 1 is a schematic diagram showing the configuration of an autonomous construction robot control system S according to an embodiment of the present invention.
The construction site KG is an area where construction work is carried out using an autonomous construction robot R. For example, the construction site KG is an underground space formed by the inverted construction method. Figure 1 shows the construction site KG as seen from above. A retaining wall YK is formed around the inner periphery of the construction site KG.

Multiple acceleration sensors KS are installed on the retaining wall YK. The acceleration sensors KS are used as retaining inclinometers to measure the inclination of the retaining wall YK. These acceleration sensors KS are, for example, MEMS (Micro Electro Mechanical Systems) sensors. Such retaining inclinometers are sometimes installed in underground spaces formed by inverted construction and used to monitor the inclination of the retaining wall. In this embodiment, the acceleration sensors KS are used not only to monitor the inclination of the retaining wall, but also to detect vibrations occurring in the target area, which is the construction site KG. By repurposing the retaining inclinometers as acceleration sensors to detect vibrations occurring within the site, installation costs are reduced because there is no need to install a separate acceleration sensor for the retaining inclinometers. In this case, the acceleration sensors KS detect vibrations occurring within the construction site KG. The vibrations detected here include vibrations caused by the operation of construction machinery used at the construction site KG. In other words, the acceleration sensors also function as vibration sensors.

The acceleration sensors KS are arranged horizontally (in the X and Y directions) at regular intervals (e.g., at intervals of several tens of meters) to surround the excavation area, and vertically at regular intervals (e.g., at intervals of 2 meters). By arranging multiple acceleration sensors KS horizontally and vertically in this manner, the position of the vibration source generated by construction machinery at the construction site KG can be estimated, and the operating status of the construction machinery can be grasped in three-dimensional space. In other words, it is possible to grasp the location at the construction site KG where the construction machinery is operating. Furthermore, if multiple acceleration sensors KS are installed at different positions horizontally to surround the excavation area (only one in the vertical direction), the position of the construction machinery can also be grasped in two-dimensional horizontal space. In this case, the number of acceleration sensors KS to be installed can be reduced compared to when multiple sensors are installed vertically.

Furthermore, there are multiple construction machines at the construction site KG. There are various types of construction machines used at the construction site KG, including backhoes, hammer grab drops, and breakers. FIG. 1 shows an example in which there are three construction machines (e.g., backhoes). Construction machine J1 and construction machine J2 are in operation and may perform operations such as swinging. Construction machine J3 is present at the construction site KG, but is turned off and is in an inoperative state.
Here, even if the construction machines J1 and J2 are in an operating state and are not traveling and their current positions do not change, when they turn, the posture of the construction machines changes, and there is a possibility that the arms may interfere with other objects in the turning range SH. The construction machine J3 is in an inoperable state and will not turn, so its posture will not change.
In this embodiment, the construction machine may be a large or medium-sized heavy machine, or may be a small construction machine as long as its position is not changed but its posture may be changed.

The autonomous construction robot R is an unmanned construction machine that can operate autonomously. The unmanned construction machine may be, for example, a backhoe, a dump truck, a bulldozer, or any other machine that can operate unmanned. If the autonomous construction robot R is a backhoe, underground excavation can be performed by the robot. If the autonomous construction robot R is a dump truck, the transportation of transported items can be made more efficient.
When a movement path (patrol route) IK is externally specified, the autonomous construction robot R moves along the movement path. Here, the autonomous construction robot R needs to patrol while avoiding interference with the retaining wall YK and with each construction machine.

FIG. 2 is a schematic functional block diagram showing the configuration of the autonomous construction robot control device 1.
The autonomous construction robot control device 1 will be described as being mounted on the autonomous construction robot R, but it may also be provided outside the autonomous construction robot R and connected to be able to communicate with the autonomous construction robot R. The autonomous construction robot control device 1 may also be configured as a server and may communicate with the autonomous construction robot R by being connected at least in part via a wireless network.

The autonomous construction robot control device 1 includes a sensor data acquisition unit 11, a frequency data storage unit 12, an analysis unit 13, an object position estimation unit 14, a self-position estimation unit 15, and a control unit 16.

The sensor data acquisition unit 11 acquires data from various sensors. Sensors from which the sensor data acquisition unit 11 can acquire data include an acceleration sensor KS, a thermography camera TH, and a positioning sensor LI.

The sensor data acquisition unit 11 acquires waveform data, which is the detection result of detected vibrations, from an acceleration sensor KS that is installed in the target area where construction work is to be carried out and detects vibrations.

The sensor data acquisition unit 11 also acquires temperature distribution data generated by the thermographic camera TH. This thermographic camera TH is provided on the autonomous construction robot R, analyzes infrared rays emitted from objects around the autonomous construction robot R, and generates temperature distribution data representing the temperature distribution from the analysis results. When the autonomous construction robot control device 1 is mounted on the autonomous construction robot R, the thermographic camera TH may be provided on the autonomous construction robot control device 1.

The sensor data acquisition unit 11 also acquires positioning data from the positioning sensor LI. Here, the positioning sensor LI may be, for example, any one of a LiDAR (Light Detection and Ranging), a ToF (Time Of Flight) sensor, a camera, a gyro sensor, etc. If the positioning sensor LI is a LiDAR, the positioning sensor LI generates point cloud data that measures distances to objects in the surrounding three-dimensional space. The positioning sensor LI is mounted on the autonomous construction robot R. Furthermore, if the autonomous construction robot control device 1 is mounted on the autonomous construction robot R, the positioning sensor LI may also be mounted on the autonomous construction robot control device 1.

The frequency data storage unit 12 stores the operating status of the construction machine and frequency characteristics corresponding to the operating status as frequency characteristic data. Fig. 3 is a diagram showing an example of frequency characteristic data stored in the frequency data storage unit 12. The frequency characteristic data includes the type of construction machine, the operating status of the construction machine, frequency, and vibration level.
The type of construction machine indicates the type of construction machine, such as a backhoe, a hammer grab, a breaker, etc.
The operational status of a construction machine indicates how the construction machine is operating. For example, the operational status of a backhoe includes swinging, raising the arm, lowering the arm, etc.
The frequency indicates the main frequency of vibrations that occur when construction machinery is in operation. For example, the main frequency that occurs when a "backhoe" is "swinging" is "f11 to f12." Such vibrations are generated by the operation of construction machinery and transmitted through the ground to the retaining wall YK, where they are detected by the acceleration sensor KS. In addition, vibrations generated by the operation of construction machinery may also be transmitted to the acceleration sensor KS through the air and detected by the acceleration sensor KS.
The main frequency of such frequencies varies depending on the construction machine and its operating conditions. Vibrations at these frequencies are measured in advance using the acceleration sensor KS for each operating condition of the construction machine. The frequency data storage unit 12 then stores the measurement results measured in advance by the acceleration sensor KS for each operating condition of the construction machine, in association with the type of construction machine and its operating condition.
The vibration level indicates the magnitude of vibrations that occur when a construction machine is operating. For example, the vibration level that occurs when a "backhoe" is "swinging" is "a11 to a12." Such vibration levels vary depending on the construction machine and its operating conditions. These vibration levels are stored for each operating condition of the machine, in association with the type of construction machine, operating conditions, and frequency, as measured in advance by the acceleration sensor KS.

The frequency data storage unit 12 is configured by a storage medium, for example, a hard disk drive (HDD), a flash memory, an electrically erasable programmable read-only memory (EEPROM), a random access read/write memory (RAM), a read-only memory (ROM), or any combination of these storage media.
The frequency data storage unit 12 can be implemented by using, for example, a nonvolatile memory.

The analysis unit 13 includes a frequency analysis unit 131 and a temperature data analysis unit 132. The analysis unit 13 performs various analyses on the data obtained from the sensor data acquisition unit 11.
The frequency analysis unit 131 analyzes frequencies included in the waveform data acquired by the sensor data acquisition unit 11 and estimates the operating status of the construction machinery operating in the target area. For example, the frequency analysis unit 131 extracts frequencies included in the waveform data by performing the analysis, and for the main frequencies among the extracted frequencies, references the frequency data storage unit 12 and reads out the operating status corresponding to the extracted frequencies, thereby estimating the operating status of the construction machinery. A characteristic frequency component may be used as the main frequency. For example, the characteristic frequency component may be a frequency component that has a certain level of acceleration or more than other frequency components.
Here, the analysis unit 13 may not only analyze the frequency using the frequency analysis unit 131, but also detect the vibration level based on the amplitude of the waveform data, and estimate the operating status of the construction machine by referring to the frequency data storage unit 12 based on a combination of the vibration level and the main frequency based on the analysis result of the frequency analysis unit 131, and reading out the type and operating status of the construction machine corresponding to this combination. This makes it possible to improve the estimation accuracy.

In this way, the frequency analysis unit 131 can grasp the operating status of construction machinery based on waveform data acquired by acceleration sensors installed within the construction site. When the vibration source is due to the work of construction machinery, the magnitude and frequency of the vibratory force at the vibration source will vary depending on the type of construction machinery and the working style. Therefore, by analyzing the frequency of the acquired waveform, it is possible to identify the operating status as well as the location and type (backhoe rotation, hammer grab drop, breaker impact, etc.).

FIG. 4 shows an example of waveform data detected by the acceleration sensor KS. In FIG. 4, the horizontal axis represents time (sec) and the vertical axis represents acceleration (G). The figure shows a case in which waveform data (reference numeral 401) measured when various construction machines are not operating (non-operating) at the construction site KG is superimposed on waveform data measured when a construction machine is operating at the construction site KG. When the construction machines are not operating at the construction site KG, the acceleration indicated by the waveform data falls within a certain level. On the other hand, when any construction machine is operating at the construction site KG, the acceleration measured is higher than when the construction machine is not operating because the acceleration caused by the operation of the construction machine is included. Therefore, a reference range (reference numeral 403) is predetermined based on the acceleration when the construction machine is not operating. The analysis unit 13 compares the waveform data with this predetermined reference range. If the acceleration included in the waveform data exceeds the reference range, it can be determined that the construction machine is operating. If the acceleration included in the waveform data does not exceed the reference range, it can be determined that the construction machine is not operating.
When the frequency analysis unit 131 determines that the acceleration contained in the waveform data exceeds the reference range, it can identify the construction machinery in operation by analyzing the frequency components contained in the vibration of the waveform data when acceleration exceeding the reference range is detected.

The temperature data analysis unit 132 detects whether or not a detection target is present based on the temperature distribution data obtained from the sensor data acquisition unit 11. For example, the temperature data analysis unit 132 analyzes the temperature distribution data to detect whether or not construction machinery is operating and whether or not there are people (workers, engineers, etc.) active at the construction site KG.
The temperature data analysis unit 132 detects the presence of a person when the temperature distribution data includes a temperature distribution corresponding to a human body temperature. This makes it possible to detect the presence of a person around the autonomous construction robot R.

If the temperature distribution data includes a temperature distribution that corresponds to the temperature when a construction machine is operating, the temperature data analysis unit 132 detects that a construction machine is in operation. This makes it possible to detect that a construction machine is in operation around the autonomous construction robot R.
When construction machinery is operating, the temperature may be higher than the ambient temperature by a certain amount (a temperature corresponding to the heat generated by the engine or the motor, etc.), and if that temperature is above a threshold value, it can be determined that the construction machinery is operating.

The object position estimation unit 14 estimates the position of the construction machine based on the analysis result of the analysis unit 13, and generates position information based on the estimation result.
For example, the object position estimation unit 14 estimates the position of a construction machine based on the analysis results of the frequency analysis unit 131. Here, the position of the construction machine at the construction site KG is estimated based on the time difference between the times when the operating status of the same construction machine is obtained from the waveform data obtained from each of the multiple acceleration sensors KS and the positions of each acceleration sensor KS. For example, the object position estimation unit 14 analyzes the frequencies contained in the waveform data obtained from each acceleration sensor KS and estimates the type and operating status of the construction machine based on the dominant frequency by referring to the frequency data storage unit 12. Then, when the same combination of type and operating status of the construction machine is estimated based on each acceleration sensor KS, the object position estimation unit 14 can estimate the position of the construction machine based on the difference in the time when the dominant frequency arrives and the relationship between the positions of each acceleration sensor KS. Here, the type and operating status of the construction machine may be estimated using not only the frequencies contained in the waveform data but also the vibration level. In this way, the object position estimation unit 14 can estimate the position of the construction machine based on the positions of the multiple acceleration sensors and the waveform data detected by each acceleration sensor.

The object position estimation unit 14 may also estimate the position of a person based on the analysis results of the temperature data analysis unit 132 .
For example, when the temperature data analysis unit 132 detects a temperature distribution corresponding to a person's body temperature, the object position estimation unit 14 extracts an area corresponding to the person's body temperature from the temperature distribution data and estimates the location of the person based on the direction and size of the area. Here, the size of the area on the image in which a person is detected is determined based on the average height of the person and the distance from the thermographic camera TH. Therefore, the object position estimation unit 14 can estimate the distance to the thermographic camera TH based on the size of the area in which the person's body temperature was detected, and can identify the location in which the person is estimated to be present by also taking into account the direction in which the area in which the person's body temperature was detected was captured.

The object position estimation unit 14 may also estimate the position of the construction machine based on the analysis results of the temperature data analysis unit 132 and the results of analysis by the temperature data analysis unit 132 .
The object position estimation unit 14 can estimate the direction of the operating construction machinery from the area (e.g., pixel position) in the temperature distribution data where the presence of the operating construction machinery is detected and the shooting direction of the thermographic camera TH.
Here, the object position estimation unit 14 extracts an area corresponding to the shape of the construction machine from the distribution of point cloud data included in the positioning data, and estimates the location of the construction machine based on the distance represented by the point cloud data and the direction in which the point cloud data was obtained. It then determines whether the location estimated as the location of the construction machine matches the direction and distribution status (area of distribution) of operating construction machines analyzed by the temperature data analysis unit 132, and if they match, it can be estimated that the construction machine is operating at the location estimated as the location of the construction machine.

The self-position estimation unit 15 estimates the position of the autonomous construction robot R controlled by the autonomous construction robot control device 1 based on the measurement results of the positioning sensor LI output from the sensor data acquisition unit 11. The self-position estimation unit 15 can estimate its own position from the measurement results obtained by the positioning sensor LI by using SLAM (Simultaneous Localization and Mapping) technology. Here, a technology that estimates its own position using SLAM using LiDAR as the positioning sensor LI is sometimes referred to as LiDAR SLAM. A technology that estimates its own position using SLAM using a Time of Flight (ToF) sensor as the positioning sensor LI is sometimes referred to as Depth SLAM. A technology that estimates its own position using SLAM using a camera as the positioning sensor LI is sometimes referred to as Visual SLAM.

The control unit 16 outputs various control signals to the autonomous construction robot R. The control unit 16 includes a route generation unit 161.
When a detection target exists around the autonomous construction robot R, the route generation unit 161 outputs a control signal to the autonomous construction robot R to operate the robot so as not to interfere with the detection target. The detection target is at least one of an operating construction machine and a person.
For example, if there is an operating construction machine based on the operation status of the construction machine, the route generation unit 161 outputs a control signal to the autonomous construction robot R so as not to interfere with the operating construction machine. If the operating construction machine is a backhoe, the control unit 16 can control the autonomous construction robot R so as not to interfere with the backhoe by controlling the autonomous construction robot R so as not to enter the turning range of the backhoe.
Furthermore, when the route generation unit 161 detects the presence of a person in the vicinity based on the analysis results of the temperature data analysis unit 132, it outputs a control signal to the autonomous construction robot R so as not to interfere with the person. When a person is present in the vicinity, the control unit 16 can control the autonomous construction robot R so as not to interfere with the person by controlling the autonomous construction robot R not to come within a certain distance from the person.

When performing control to avoid interference, the route generation unit 161 generates a travel route that does not interfere with the estimated position of the operating construction machine or the detected person, depending on the operating status, and outputs a control signal representing the generated travel route.
Furthermore, when controlling to avoid interference, the control unit 16 may avoid interference by temporarily suspending the movement or operation (turning, up and down movement of the arm, etc.) of the autonomous construction robot control device 1, rather than generating a movement path that does not interfere with operating construction machinery or people. For example, when an operating construction machinery or person passes near the autonomous construction robot R, the autonomous construction robot R may continue moving by canceling the pause after the construction machinery or person has passed from the movement path of the autonomous construction robot R. This allows the autonomous construction robot R to move while avoiding interference with operating construction machinery (moving construction machinery) or people. Here, in addition to temporarily suspending to avoid interference, the autonomous construction robot R may also slow down its movement speed, and temporarily suspend its movement if the presence of operating construction machinery or people is continuously detected.

The above-mentioned sensor data acquisition unit 11, analysis unit 13, object position estimation unit 14, self-position estimation unit 15, and control unit 16 may be configured as a processing unit such as a CPU (central processing unit) or a dedicated electronic circuit.

FIG. 5 is a flowchart illustrating the operation of the autonomous construction robot control device 1.
The autonomous construction robot control device 1 is mounted on, for example, the autonomous construction robot R, and executes processing in response to the autonomous construction robot R being powered on.
The sensor data acquisition unit 11 of the autonomous construction robot control device 1 acquires measurement results from each sensor (step S101).
When the measurement results are acquired by the sensor data acquisition unit 11, the analysis unit 13 performs an analysis based on the measurement results obtained by the sensor data acquisition unit 11. Here, the analysis unit 13 analyzes whether or not there are people around the autonomous construction robot R and whether or not there is construction machinery in operation within the construction site KG (step S102).
If the analysis unit 13 detects that there are no people around the autonomous construction robot R and that there are no construction machines in operation within the construction site KG, it transitions the processing to step 101 (step S103-NO).

On the other hand, if the analysis unit 13 detects at least one of the presence of a person around the autonomous construction robot R and the presence of operating construction machinery within the construction site KG (step S103-YES), it notifies the object position estimation unit 14 that a detection target has been detected. The object position estimation unit 14 estimates the position of the detected detection target (step S104). For example, if the analysis results of the analysis unit 13 detect a person, the object position estimation unit 14 estimates the position of the person, and if the analysis results of the analysis unit 13 detect the presence of operating construction machinery, it estimates the position of the operating construction machinery.

Meanwhile, the self-position estimation unit 15 acquires data measured by LiDAR from the sensor data acquisition unit 11 at regular intervals and estimates the self-position of the autonomous construction robot R.

When the object position estimation unit 14 estimates the position of the detection target, the route generation unit 161 of the control unit 16 generates a movement route that does not interfere with the detection target (step S105) and outputs the generated movement route to the autonomous construction robot R (step S106). The autonomous construction robot R moves based on the movement route output from the control unit 16.
On the other hand, the control unit 16 determines whether or not to terminate the operation of the autonomous construction robot R (step S107), and if the operation of the autonomous construction robot R is to be terminated, terminates the processing of the autonomous construction robot control device 1 (step S107-YES), and if the operation of the autonomous construction robot R is not to be terminated, transitions the processing to step S101.

According to the embodiment described above, by analyzing waveform data acquired by multiple acceleration sensors (MEMS sensors) installed in the retaining wall, it is possible to instantly and accurately determine the operating status, location, and type of construction machinery in underground spaces using the inverted construction method.
Furthermore, according to the embodiment described above, in addition to estimating the robot's own position using conventional SLAM technology, the operating status of construction machinery can be grasped based on waveform data acquired by acceleration sensors installed within the construction site. If an operating construction machinery is detected, a movement path is regenerated. The autonomous construction robot R can safely patrol the construction site while appropriately reconfiguring its movement path to avoid areas where it has determined that construction machinery is operating.
Furthermore, in the above-described embodiment, an autonomous robot equipped with SLAM technology can be safely and inexpensively operated in narrow spaces where construction machinery can operate, such as underground spaces used in inverted construction. Because it is easy to introduce autonomous robots into such underground construction work, the autonomous construction robot control device 1 described above contributes to improving the efficiency and productivity of management work.

Furthermore, according to the above-described embodiment, it is possible to grasp the status of whether construction machinery is operating, its operating location, etc., and by grasping this operating status, it is also possible to check the progress of construction work in underground spaces.

Furthermore, for example, in the above-described embodiment, if it is detected that construction machinery is operating in the target area, the autonomous construction robot control device 1 can temporarily stop the autonomous construction robot, or generate a movement route that bypasses the target area and send it to the autonomous construction robot. As a result, if construction machinery is operating in the target area, the autonomous construction robot can temporarily stop, and if it is detected that the construction machinery is operating in an operating state where its posture does not change (for example, if its posture does not change for more than a certain period of time), it can resume movement.

Furthermore, if it is detected that the construction machine is in an operating state where its posture does not change in the target area (such as when it is not operating, or when the construction machine's lights are on but no operations involving a change in posture are being performed), it may move along a pre-generated movement path, and if a construction machine is operating in the target area, it may regenerate a movement path that detours around the area where the construction machine is operating. This allows the autonomous construction robot to move along a pre-determined movement path when the construction machine's posture does not change in the target area, and to detour around the construction machine when a construction machine is operating in the target area so as not to interfere with it.

In the above-described embodiment, the case where the frequency analysis unit 131 estimates the operating status by referring to the frequency data storage unit 12 has been described. However, the frequency analysis unit 131 may estimate the operating status without using the frequency data storage unit 12.
For example, data representing the operating status of the construction machine may be assigned to the waveform data as a teacher label, and a learning device may be used to learn the relationship between the waveform data and the operating status of the construction machine, thereby generating a trained model. Machine learning, deep learning, or the like may be used as a learning method in the learning device. The frequency analysis unit 131 may then input the waveform data obtained from the sensor data acquisition unit 11 into the trained model obtained from the learning device to obtain the operating status of the construction machine.

In addition, in the above-described embodiment, the construction site KG was described as an underground space formed by the inverted construction method, but the construction site KG may be an underground space other than an underground space formed by the inverted construction method, as long as it is a construction site where an autonomous construction robot R is introduced and where an acceleration sensor KS can be installed. For example, the construction site KG may be the interior of a tunnel.

The autonomous construction robot control device 1 in the above-described embodiment may be implemented by a computer. In this case, a program for implementing this function may be recorded on a computer-readable recording medium, and the program may be loaded into a computer system and executed. Note that the term "computer system" here includes hardware such as an OS and peripheral devices. Furthermore, "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into computer systems. Furthermore, "computer-readable recording medium" may also include devices that dynamically store programs for a short period of time, such as communication lines used when transmitting programs over networks such as the Internet or telephone lines, or devices that store programs for a fixed period of time, such as volatile memory within the computer system that serves as the server or client. The program may also be designed to implement some of the above-described functions, or may be capable of implementing the above-described functions in combination with programs already stored in the computer system, or may be implemented using a programmable logic device such as an FPGA (Field Programmable Gate Array).

The above describes in detail an embodiment of the present invention with reference to the drawings, but the specific configuration is not limited to this embodiment and includes designs that do not deviate from the gist of the present invention.

1...Autonomous construction robot control device, 11...Sensor data acquisition unit, 12...Frequency data storage unit, 13...Analysis unit, 14...Object position estimation unit, 15...Self-position estimation unit, 16...Control unit, 101...Step, 131...Frequency analysis unit, 132...Temperature data analysis unit, 161...Route generation unit

Claims (6)

a sensor data acquisition unit that acquires waveform data from a sensor that is installed in a target area where construction work is to be performed and detects vibrations;
an analysis unit that analyzes frequencies included in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machines operating in the target area;
a control unit that outputs a control signal to the autonomous construction robot based on the operating status of the construction machine so as to operate the autonomous construction robot without interfering with the construction machine;
an object position estimation unit that generates position information that estimates the position of the construction machine based on the analysis result of the analysis unit;
the control unit generates a travel route that does not interfere with the estimated position in accordance with the operation status, and outputs a control signal representing the generated travel route.
Autonomous construction robot control device. a frequency data storage unit that stores the operating status of the construction machine and frequency characteristics corresponding to the operating status;
The autonomous construction robot control device according to claim 1 , wherein the analysis unit estimates the operating status by referring to the frequency data storage unit and acquiring the operating status corresponding to the frequency obtained by the analysis. the sensor data acquisition unit acquires temperature distribution data obtained from a thermography camera that generates temperature distribution data representing a temperature distribution based on infrared rays emitted from objects present around the autonomous construction robot;
The analysis unit detects whether or not a detection target is present based on the temperature distribution data,
The autonomous construction robot control device according to claim 1 or 2, wherein the control unit outputs a control signal to the autonomous construction robot when the detection target is present so as not to interfere with the detection target. the sensors are a plurality of acceleration sensors that detect accelerations based on the vibrations and are provided at different positions surrounding the target area;
The autonomous construction robot control device according to claim 1 , wherein the object position estimation unit estimates the position of the construction machine based on the positions where the plurality of acceleration sensors are installed and the waveform data detected by each of the plurality of acceleration sensors. A sensor that is installed in the target area where construction work is to be performed and detects vibrations;
a sensor data acquisition unit that acquires waveform data from the sensor;
an analysis unit that analyzes frequencies included in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machines operating in the target area;
a control unit that outputs a control signal to the autonomous construction robot based on the operating status of the construction machine so as not to interfere with the construction machine;
an object position estimation unit that generates position information that estimates the position of the construction machine based on the analysis result of the analysis unit;
The control unit generates a travel route that does not interfere with the estimated position in accordance with the operating status, and outputs a control signal representing the generated travel route.
An autonomous construction robot control system having: 1. A computer-implemented method for controlling an autonomous construction robot, comprising:
The sensor data acquisition unit acquires waveform data from a sensor that is installed in the target area where the construction work is to be performed and detects vibrations.
an analysis unit that analyzes frequencies contained in the waveform data acquired by the sensor data acquisition unit and estimates the operating status of construction machines operating in the target area;
a control unit that outputs a control signal to the autonomous construction robot based on the operating status of the construction machine so as to operate the autonomous construction robot without interfering with the construction machine;
an object position estimation unit generates position information that estimates the position of the construction machine based on the analysis result of the analysis unit;
The control unit generates a travel route that does not interfere with the estimated position in accordance with the operating status, and outputs a control signal representing the generated travel route.
A method for controlling an autonomous construction robot.