Asphalt finisher

Abstract

To provide an asphalt finisher with which state of paving material construction scope can be easily apprehended.SOLUTION: An asphalt finisher includes a tractor, a hopper installed on the front side of the tractor, a conveyor that conveys paving materials in the hopper to the rear side of the tractor, a screw that spreads the paving materials conveyed by the conveyor and scattered on a road surface in a vehicle width direction, a screed that levels the paving materials spread by the screw on the rear side of the screw, and a detection part that detects a construction area of the paving materials, and is configured to perform processing to output information regarding a status of a person detected by the detection part.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to an asphalt finisher.

Background Art

[0002] Conventionally, an asphalt finisher has been known that includes a tractor, a hopper installed in front of the tractor for receiving paving material, a conveyor for conveying the paving material in the hopper to the rear side of the tractor, a screw for spreading the paving material conveyed by the conveyor in the vehicle width direction at the rear side of the tractor, and a screed for leveling the paving material spread by the screw at the rear side of the screw.

[0003] For example, Patent Document 1 describes a paving machine having a conveyor entrapment prevention function that includes a hopper provided at the front part of a vehicle body and a conveyor for conveying the paving material accommodated in the hopper to the rear of the vehicle body, and prohibits the operation of the conveyor when a person inside the hopper is detected.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the invention described in Patent Document 1 only detects a person inside the hopper and does not detect a person present around the vehicle body.

[0006] One aspect of the present invention aims to provide an asphalt finisher capable of easily knowing the situation of the construction range of paving material.

Means for Solving the Problems

[0007] An asphalt finisher according to one aspect of the present invention includes a tractor, a hopper installed on the front side of the tractor to receive paving material, a conveyor that conveys the paving material in the hopper to the rear side of the tractor, a screw that spreads the paving material conveyed by the conveyor in the vehicle width direction on the rear side of the tractor, a screed that levels the paving material spread by the screw on the rear side of the screw, and a detection unit that detects the construction range of the paving material, and is configured to perform a process of outputting information regarding the state of a person detected by the detection unit.

Effects of the Invention

[0008] According to one aspect of the present invention, the situation of the construction range of the paving material can be easily known.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Modes for Carrying Out the Invention

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Also, the embodiments described below are illustrative rather than limiting the invention, and not all features and combinations thereof described in the embodiments are necessarily essential to the invention. In each drawing, the same or corresponding components are denoted by the same or corresponding reference numerals, and the description may be omitted.

[0011] [Embodiment] One embodiment of the present invention is a road paving system for paving a road surface using an asphalt finisher. The asphalt finisher according to this embodiment has a function of detecting the state of a person existing around it.

[0012] At a work site where road paving and the like are carried out, many workers work around the asphalt finisher. The workers need to change the working place as the asphalt finisher moves. At such a work site, if an abnormality occurs due to a worker's fall or the like, or if the worker's attention is not directed towards the asphalt finisher, there is a risk of an abnormality such as the worker coming into contact with the asphalt finisher body or the screed.

[0013] In order to suppress such an abnormality, it is conceivable that the asphalt finisher is provided with a function of stopping when a person is detected. However, usually, since workers work while paying attention to the operation near the asphalt finisher, the asphalt finisher will frequently stop operating due to the detection of workers, which will affect the progress of the work. Therefore, the asphalt finisher is required to have a function of being able to stop the progress of the tractor and the rotation of the screed according to a change in the state of a person existing around it.

[0014] <Overall Configuration of the Road Paving System> The overall configuration of the road paving system SYS according to the embodiment will be described with reference to FIG. 1. FIG. 1 is a schematic diagram showing an example of the overall configuration of the road paving system according to the embodiment.

[0015] As shown in FIG. 1, the road pavement system SYS according to the embodiment includes an asphalt finisher 100, a communication terminal 200 (an example of an external device), and a remote management device 300 (an example of an external device). The asphalt finisher 100 and the remote management device 300 are connected by a public network NT.

[0016] Further, in the communication terminal 200, for example, the road pavement system SYS performs various settings related to the control of the asphalt finisher 100 in response to an input from a user or automatically, and transmits them to the asphalt finisher 100. Thereby, various operations of the asphalt finisher 100 can be controlled and monitored from the communication terminal 200.

[0017] Also, the asphalt finisher 100 may transmit information indicating the current situation to one or more of the communication terminal 200 and the remote management device 300. Furthermore, the asphalt finisher 100 may transmit log information indicating the detection results of the surroundings to one or more of the communication terminal 200 and the remote management device 300.

[0018] The remote management device 300 is a terminal provided for remotely managing the work site. The remote management device 300 is operated by a user or the like who remotely manages the work site. For example, the remote management device 300 receives image information indicating the detection results around the asphalt finisher 100 from the asphalt finisher 100 and displays it on a display device (liquid crystal panel) (not shown). Also, the remote management device 300 can manage the situation of the past work site by storing the log information transmitted from the asphalt finisher 100.

[0019] The communication terminal 200 is, for example, a terminal possessed by a user who manages work at a work site, or a user who is working at a work site. The communication terminal 200 receives image information indicating the detection results around the asphalt finisher 100 from the asphalt finisher 100 and displays it on a display device (liquid crystal panel) (not shown). Thereby, the user who manages work at the work site can recognize the detection results around the asphalt finisher 100.

[0020] There may be one or a plurality of communication terminals 200 included in the road paving system SYS. Thereby, the road paving system SYS can provide information regarding the asphalt finisher 100 to a plurality of users who each use it through the plurality of communication terminals 200.

[0021] There may be one or a plurality of asphalt finishers 100 included in the road paving system SYS. Thereby, the road paving system SYS can perform data collection, information provision to users based on the collected data, settings related to the control of the asphalt finisher 100, etc. with respect to the asphalt finisher 100.

[0022] <Overview of Asphalt Finisher> The overview of the asphalt finisher 100 according to the embodiment will be described with reference to FIGS. 2 to 4. FIG. 2 is a side view showing an example of the asphalt finisher according to the embodiment. FIG. 3 is a top view showing an example of the asphalt finisher according to the embodiment. FIG. 4 is a rear view showing an example of the asphalt finisher according to the embodiment.

[0023] The asphalt finisher 100 mainly includes a tractor 1, a hopper 2, and a screed 3.

[0024] The tractor 1 is a device for driving the asphalt finisher 100 and pulls the screed 3. In this embodiment, the tractor 1 rotates two or four wheels using a traveling hydraulic motor to move the asphalt finisher 100. The traveling hydraulic motor rotates by receiving the supply of hydraulic oil from a hydraulic pump driven by a prime mover such as a diesel engine. A driver's seat 1S and an operation panel 65 are arranged on the upper part of the tractor 1.

[0025] An imaging device 51 (an example of a detection unit) is attached to the tractor 1. The imaging device 51 includes a right camera 51R attached to the right side, a left camera 51L attached to the left side, and a front camera 51F attached to the front. A display device 52 is installed at a position where it is easily visible to the driver sitting on the driver's seat 1S. In this embodiment, the direction of the hopper 2 as seen from the tractor 1 is defined as the front (+X direction), and the direction of the screed 3 as seen from the tractor 1 is defined as the rear (-X direction). The +Y direction corresponds to the left direction, and the -Y direction corresponds to the right direction.

[0026] The hopper 2 is a mechanism for receiving paving materials (for example, asphalt mixture). The hopper 2 is a device for supplying paving materials in front of the screed 3. In this embodiment, the hopper 2 is configured to be openable and closable in the vehicle width direction by a hydraulic cylinder. The asphalt finisher 100 usually receives paving materials from the loading platform of the dump truck with the hopper 2 fully open. Then, when the paving materials in the hopper 2 decrease, the hopper 2 is closed, and the paving materials near the inner wall of the hopper 2 are collected at the center of the hopper 2 so that the conveyor CV can convey the paving materials to the screed 3.

[0027] The conveyor CV is driven by a hydraulic motor that rotates by receiving the supply of hydraulic oil from a hydraulic pump. In this embodiment, the conveyor CV is configured to send the paving materials in the hopper 2 to the rear side of the tractor 1 through a conveying passage. The conveying passage is a substantially rectangular parallelepiped-shaped space formed inside the tractor 1 and has a substantially rectangular inlet that opens into the hopper 2 on the front surface of the tractor 1.

[0028] The screw SC is driven by a hydraulic motor that rotates upon receiving the supply of hydraulic oil from a hydraulic pump. In this embodiment, the screw SC includes a central screw (not shown), a left screw, and a right screw. The central screw is installed within the width of the tractor 1. The left screw is connected to the left end of the central screw and is installed so as to protrude to the left from the width of the tractor 1. The right screw is connected to the right end of the central screw and is installed so as to protrude to the right from the width of the tractor 1.

[0029] The screed 3 is a mechanism for leveling the paving material. In this embodiment, it is configured to be vertically movable and horizontally expandable and contractible by a hydraulic cylinder. The width of the screed 3, when expanded in the vehicle width direction, is larger than the width of the tractor 1. In this embodiment, the screed 3 includes a main screed 30, a left telescopic screed 31L, and a right telescopic screed 31R. The left telescopic screed 31L and the right telescopic screed 31R are configured to be expandable and contractible in the vehicle width direction (Y-axis direction). And the left telescopic screed 31L and the right telescopic screed 31R that are expandable and contractible in the vehicle width direction are arranged offset from each other in the traveling direction (X-axis direction). Therefore, it can have a longer width (length in the vehicle width direction) than when not offset, can be extended longer in the vehicle width direction, and a wider newly constructed paving body can be constructed.

[0030] The controller 50 is a control unit that controls the asphalt finisher 100. The controller 50 is, for example, a computer equipped with a CPU (Central Processing Unit), a volatile memory, a non-volatile memory, etc. The controller 50 is a computer including a CPU and a RAM (Random Access Memory), and is mounted on the tractor 1. Various functions of the controller 50 are realized, for example, by the CPU executing a program stored in the auxiliary storage device 48.

[0031] The auxiliary storage device 48 is a device for storing various information. In the present embodiment, the auxiliary storage device 48 is a non-volatile memory and is integrated with the controller 50. However, the auxiliary storage device 48 may be arranged outside the controller 50 as a separate structure from the controller 50.

[0032] The imaging device 51 is attached to the tractor 1. The imaging device 51 is configured to acquire information regarding the space around the asphalt finisher 100 and output the acquired information to the controller 50. The imaging device 51 according to the present embodiment includes a front camera 51F, a left camera 51L, and a right camera 51R. The imaging device 51 may be attached at a position other than the right side, left side, and front of the tractor 1 (for example, the rear). The imaging device 51 may be equipped with a wide-angle lens or a fish-eye lens. The imaging device 51 may be attached to the hopper 2 or the screed 3.

[0033] The imaging device 51 according to the present embodiment is, for example, a camera equipped with an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal-Oxide-Semiconductor). The imaging device 51 may be any space recognition device that can recognize the space based on the asphalt finisher 100. For example, LiDAR (Light Detection and Ranging) may be used.

[0034] As shown in FIGS. 2 and 3, the front camera 51F is attached to the upper end of the front of the tractor 1, and its optical axis 51FX extends forward in the traveling direction and is attached so as to form an angle α in a side view with the road surface. As shown in FIGS. 2 to 4, the left camera 51L is attached to the upper end of the left side of the tractor 1, and its optical axis 51LX forms an angle β in a top view with the left side surface of the tractor 1 and forms an angle γ in a rear view with the road surface. The right camera 51R is attached in the same manner as the left camera 51L with the left and right reversed.

[0035] The area 51FA enclosed by the dashed line in FIG. 3 indicates the imaging range of the front camera 51F, the area 51LA enclosed by the dashed-dotted line indicates the imaging range of the left camera 51L, and the area 51RA enclosed by the dashed-dotted line indicates the imaging range of the right camera 51R. As shown in FIG. 3, the range that can be imaged by the imaging device 51 includes the traveling direction of the tractor 1, the side of the hopper 2, the front of the screed 3, and the front of the operating range of the screw SC.

[0036] The imaging device 51 is attached to the asphalt finisher 100 via, for example, brackets, stays, bars, etc. In the present embodiment, the imaging device 51 is attached to the tractor 1 via an attachment stay. However, the imaging device 51 may be directly attached to the tractor 1 without using the attachment stay, or may be embedded in the tractor 1.

[0037] In the present embodiment, the imaging device 51 outputs the acquired input image to the controller 50. When the imaging device 51 acquires the input image using a fish-eye lens or a wide-angle lens, it may output to the controller 50 a corrected input image in which the apparent distortion and fish-eye effect caused by using those lenses are corrected. Alternatively, it may output the input image without correcting the apparent distortion and fish-eye effect as it is to the controller 50. In this case, the apparent distortion and fish-eye effect are corrected by the controller 50.

[0038] The display device 52 is a device for displaying various information. In the present embodiment, the display device 52 is a liquid crystal display installed on the operation panel 65, and displays various images output by the controller 50.

[0039] The retaining plate 70 is a plate-like member for preventing the paving material fed in the vehicle width direction by the screw SC from scattering in front of the screw SC so that the paving material can be appropriately fed in the vehicle width direction by the screw SC.

[0040] The side plate 71 is also attached to the distal end of the mold board 72. The mold board 72 is a member for adjusting the amount of paving material that stays in front of the left telescopic screed 31L and the right telescopic screed 31R among the paving materials spread by the screw SC, and is configured to be able to expand and contract in the vehicle width direction together with the left telescopic screed 31L and the right telescopic screed 31R.

[0041] The controller 50 mounted on the asphalt finisher 100 will be described with reference to FIG. 5. FIG. 5 is a block diagram showing an example of the controller of the asphalt finisher according to the embodiment and the devices connected to the controller.

[0042] As shown in FIG. 5, an auxiliary storage device 48, an imaging device 51, a display device 52, a communication device 53, a drive system controller 54, a screw control device 55, and an audio output device 56 are connected to the controller 50.

[0043] The auxiliary storage device 48 stores a log information storage unit 48a and a learned model LM. The log information storage unit 48a stores log information that is the detection result around the asphalt finisher 100. The log information will be described later. The learned model LM is a machine learning model learned to detect a person from an input image. The learned model LM will be described later.

[0044] The communication device 53 performs wireless communication with devices existing around the asphalt finisher 100 or a server that manages the work site. For example, the communication device 53 performs wireless communication using any one or more of Wi-Fi (registered trademark), wireless LAN, and Bluetooth (registered trademark) as the wireless communication standard.

[0045] The drive system controller 54 controls the tractor 1 according to a control command. For example, the drive system controller 54 performs rotational control (speed control) on the rear wheel driving motor of the tractor 1 and steering angle control on the front wheels (an example of drive wheels) of the tractor 1 so as to follow the steering angle and speed indicated by the control command.

[0046] The screw control device 55 is configured to control the rotational speed of the screw SC. For example, the screw control device 55 is a solenoid valve that controls the flow rate of the hydraulic oil flowing into the hydraulic motor that drives the screw SC. Specifically, the screw control device 55 increases or decreases the flow passage area of the pipeline connecting the hydraulic motor that drives the screw SC and the hydraulic pump according to a control command from the controller 50. More specifically, the screw control device 55 increases the flow rate of the hydraulic oil flowing into the hydraulic motor that drives the screw SC by increasing the flow passage area, thereby increasing the rotational speed of the screw SC. Alternatively, the screw control device 55 decreases the flow rate of the hydraulic oil flowing into the hydraulic motor that drives the screw SC by reducing the flow passage area, thereby decreasing the rotational speed of the screw SC.

[0047] The voice output device 56 is a device that outputs voice toward the surroundings of the asphalt finisher 100. For example, the voice output device 56 is a speaker that outputs voice toward the front of the asphalt finisher 100 and can output an alarm according to a command from the controller 50. Note that the voice output device 56 may output a voice message.

[0048] More specifically, the controller 50 includes an acquisition unit 101, a person detection unit 102, a state determination unit 103, a reporting unit 104, a display control unit 105, a communication control unit 106, a movement control unit 107, and a screw control unit 108 as functional blocks configured by software, hardware, or a combination thereof.

[0049] The acquisition unit 101 acquires detection information from various sensors provided in the asphalt finisher 100. For example, the acquisition unit 101 acquires image information captured by the imaging devices 51 (front camera 51F, left camera 51L, and right camera 51R).

[0050] The person detection unit 102 detects persons existing around the asphalt finisher 100 based on the detection information acquired by the acquisition unit 101. In the present embodiment, the area around the asphalt finisher 100 is the construction area for paving materials within the range that can be imaged by the imaging device 51. In the present embodiment, the construction area for paving materials includes, for example, any one or more of within a predetermined distance in the traveling direction of the asphalt finisher 100, in front of the operating range of the screw SC, within a predetermined distance on the side of the screed 3, and within a predetermined distance on the side of the hopper 2.

[0051] The predetermined distance in the traveling direction of the asphalt finisher 100 is determined according to the traveling speed of the asphalt finisher 100. The predetermined distance in the traveling direction may be, for example, about 5 to 8 meters in front of the asphalt finisher 100. The predetermined distance on the side of the screed 3 is the range in which the left telescopic screed 31L and the right telescopic screed 31R of the screed 3 can extend. The predetermined distance on the side of the hopper 2 is the range in the vehicle width direction when the hopper 2 is in the fully open state.

[0052] The person detection unit 102 detects persons from the input image (image information captured by the imaging device 51) by simply applying image processing techniques such as shape detection and pattern recognition (template matching).

[0053] In addition, the person detection unit 102 detects a person from the input image by applying machine learning in addition to, for example, image processing technology. Specifically, the person detection unit 102 uses the learned model LM that has been machine-learned to detect the characteristics of the person reflected in the input image, and outputs a rectangular frame (hereinafter, "detection frame") representing the area where the person is reflected and a label representing the state of the person from the input image. The label includes, for example, a label indicating the absence of a person and labels set for each state of the person. Only one label for each state of the person may be set, or a plurality of labels may be set. Further, when there are a plurality of persons in different states, a single learned model LM may be configured to output labels for all states, that is, to be able to detect all states, or a plurality of learned models LM that can detect only some of all states may be provided. For example, there may be a learned model LM for each type of person, and the label for each learned model LM may be composed of only a label indicating the presence of a person in a certain state and a label indicating the absence of a person in that state.

[0054] The learned model LM is generated by applying supervised learning to the base learning model. Specifically, the learned model LM is generated by causing the base learning model to perform machine learning on a collection of teacher data (teacher dataset) consisting of a combination of an image as input and a correct answer (detection frame and label) as output. Further, the learned model LM may be generated (updated) by additionally training an existing learned model LM with a new teacher dataset.

[0055] Among the images included in the teacher dataset as inputs, both images that include (depict) a person and images that do not include (depict) a person are adopted. The images that include a person include a plurality of images in which a person is depicted at different positions within the range that can be imaged by the imaging device 51, and a plurality of images in which a person is depicted in different states. The learned model LM is machine-learned by a teacher dataset that includes images in which a person is depicted at different positions, so that a person at different positions can be detected from the input image. Also, the learned model LM is machine-learned by a teacher dataset that includes images in which a person is depicted in different states, so that a person in different states can be detected from the input image.

[0056] The learned model LM is generated by an external device such as the remote management device 300, for example, and written from a predetermined recording medium to the auxiliary storage device 48 during the manufacture of the asphalt finisher 100. Also, the learned model LM may be downloaded from an external device such as the remote management device 300 to the asphalt finisher 100 through a predetermined communication line and registered in the auxiliary storage device 48. Thereby, the person detection unit 102 can detect a person using the learned model LM registered in the auxiliary storage device 48.

[0057] Also, the learned model LM may be updated by installing update data from a predetermined recording medium in the auxiliary storage device 48. Also, the learned model LM may be updated by downloading update data from an external device such as the remote management device 300 to the asphalt finisher 100 through a predetermined communication line and installing it in the auxiliary storage device 48. Thereby, the person detection unit 102 can detect a person using the updated and latest learned model LM.

[0058] For example, the person detection unit 102 detects a person from an input image by using a support vector machine (SVM) that has been machine-learned about the tendency of the image feature amounts of the persons shown in the image. In this case, the learned model LM includes, as preprocessing, a processing unit that extracts image feature amounts from the image information captured by the imaging device 51. The image feature amounts are, for example, HOG (Histogram of Oriented Gradients) feature amounts.

[0059] Also, for example, the person detection unit 102 detects a person from an input image by using a machine-learned model LM obtained by machine learning using a deep neural network (DNN), that is, deep learning (deep neural learning). Specifically, the person detection unit 102 detects a person from an input image by using a learned model LM obtained by deep learning using a convolutional neural network (CNN). The CNN is configured by connecting a plurality of combinations of a convolutional layer that performs convolutional processing and a pooling layer that performs pooling processing with an activation function, and the final determination based on the feature amounts (feature maps) is performed by the fully-connected layer in the final stage. The activation function is, for example, ReLU (Rectified Linear Unit). Thereby, the learned model LM can handle the input image as it is without requiring preprocessing.

[0060] For example, the person detection unit 102 generates regions of candidates for persons from an input image by using a learned model LM based on CNN, and classifies these candidates into labels, thereby detecting a person. That is, the learned model LM based on CNN may be, for example, a classification model that treats the detection of a person from an input image as a classification problem, generates regions of candidates for persons from the image information captured by the imaging device 51, and classifies these candidates into labels. The classification model is, for example, R (Region-based)-CNN or its derivatives (Fast R-CNN, Faster R-CNN, etc.).

[0061] Further, for example, the person detection unit 102 detects a person by simultaneously recognizing a person and specifying the position (area) thereof from an input image using a learned model LM based on CNN. That is, the learned model LM based on CNN may be a regression model that treats the detection of a person from image information captured by the imaging device 51 as a regression problem and simultaneously recognizes a person and specifies the position (area) from the input image. The regression model may be, for example, YOLO (You Only Look Once), SSD (Single Shot Detector), or the like.

[0062] The learned model LM is generated, for example, by machine learning a base learning model or an existing learned model LM so as to be able to detect monitoring objects in different postures.

[0063] The state determination unit 103 determines the state of the person detected by the person detection unit 102. When the imaging device 51 is a camera, the state determination unit 103 detects the skeleton of the person by image recognition and applies a predetermined rule to the shape of the skeleton to determine the state of the person. Specifically, when the shape of the detected skeleton indicates that the position of the knee is higher than the position of the hip, the state determination unit 103 can determine that the person has fallen.

[0064] Further, for example, the state determination unit 103 recognizes each part (eyes, nose, mouth, ears, etc.) of the person's face by image recognition and estimates the line of sight of the person based on the positional relationship of each part. Then, the state determination unit 103 can determine whether or not the line of sight of the person is directed toward the asphalt finisher 100.

[0065] Also, for example, the state determination unit 103 determines the state of a person by detecting the distance to the person through image analysis. When the imaging device 51 is a LiDAR, the state determination unit 103 detects the distance to each part of the person based on the distance data, and applies a predetermined rule to the distance to each part to determine the state of the person. Specifically, when the distance to each part of the person indicates that the legs are closer than the head, the state determination unit 103 can determine that the person has fallen.

[0066] The state determination unit 103 determines the state of the person detected from the input image by applying machine learning in addition to, for example, image processing technology. When the person detection unit 102 detects the state of the person simultaneously with the detection of the person using the learned model LM, the state determination unit 103 outputs the state detected by the learned model LM as the determination result of the state of the person.

[0067] In the present embodiment, the state determination unit 103 stores log information including the detection result indicating the person detected from the detection information and the determination result of the state of each person in the log information storage unit 48a. The log information may include the detection information acquired by the acquisition unit 101. The log information storage unit 48a accumulates the log information in time series based on the date and time when the detection information was acquired.

[0068] The reporting unit 104 determines whether a person in a state satisfying a predetermined reporting condition is detected based on the determination result by the state determination unit 103. The reporting condition is a condition for determining whether to report that the detected person is in a dangerous state or has a high possibility of being in a dangerous state. Specifically, the reporting condition includes any one or more of the following: the detected person is not standing upright, the detected person has fallen, and the line of sight of the detected person is not directed towards the asphalt finisher 100. Not standing upright includes, for example, sitting, squatting, kneeling, crawling, etc.

[0069] When it is determined that a person in a state satisfying the reporting condition has been detected, the reporting unit 104 outputs a signal for reporting that a person in a state satisfying the reporting condition has been detected. For example, the reporting unit 104 outputs a signal instructing the display device 52 to display that a person in a state satisfying the reporting condition has been detected. Also, for example, the reporting unit 104 outputs a signal instructing the audio output device 56 to emit a sound or voice indicating that a person in a state satisfying the reporting condition has been detected. The reporting unit 104 may report that a person in a state satisfying the reporting condition has been detected by turning on a warning light mounted on the asphalt finisher 100. Thereby, a person at the work site can immediately recognize that there is a person in a dangerous state around the asphalt finisher 100.

[0070] The display control unit 105 performs control to display the detection information acquired by the acquisition unit 101 on the display device 52. In the present embodiment, the display control unit 105 displays information regarding the state of the person determined by the state determination unit 103 together with the detection information acquired by the acquisition unit 101. The display control unit 105 may display only the state of the person determined by the reporting unit 104 to satisfy the reporting condition. At this time, the display control unit 105 may display a warning indicating that there is a person determined by the reporting unit 104 to satisfy the reporting condition on the display device 52. Thereby, the driver of the asphalt finisher 100 can immediately recognize that there is a person in a dangerous state around the asphalt finisher 100.

[0071] FIG. 6 is a diagram showing an example of detection information displayed on the display device of the asphalt finisher according to the embodiment. The example shown in FIG. 6 is image information 600 captured by the front camera 51F of the asphalt finisher 100, in which persons 601 and 602 determined by the reporting unit 104 to satisfy the reporting conditions are shown. As shown in FIG. 6, the person 601 captured in the image information 600 is lying on the road surface and is determined to be in a state of not standing upright or having fallen. Also, the person 602 captured in the image information 600 is determined to be in a state of not facing the asphalt finisher 100 with their line of sight.

[0072] The communication control unit 106 controls the transmission and reception of information to and from an external device using the communication device 53. In the present embodiment, the communication control unit 106 transmits and receives information to and from the communication terminal 200 or the remote management device 300.

[0073] For example, the communication control unit 106 transmits the detection information acquired by the acquisition unit 101 to the communication terminal 200 or the remote management device 300. In the present embodiment, the communication control unit 106 transmits, together with the detection information acquired by the acquisition unit 101, information regarding the state of the person determined by the state determination unit 103 to the communication terminal 200 or the remote management device 300. The communication terminal 200 or the remote management device 300 that has received the detection information and the determination result displays the detection information indicating the state of the person on the display device. Thereby, the user of the communication terminal 200 can recognize that there is a person in a dangerous state around the asphalt finisher 100 at the work site. Also, the user of the remote management device 300 can recognize that there is a person in a dangerous state around the asphalt finisher 100 from a remote location.

[0074] The movement control unit 107 outputs a control command to control the operation of the tractor 1 to the drive system controller 54. In this embodiment, when the result of the determination by the state determination unit 103 satisfies a predetermined stop condition (hereinafter referred to as a tractor stop condition), the movement control unit 107 outputs a control command to stop the progress of the tractor 1 (hereinafter referred to as a tractor stop command) to the drive system controller 54. This allows the asphalt finisher 100 to perform control to stop the progress of the asphalt finisher 100 when the state of a person present within the construction area of ​​the paving material satisfies the tractor stop condition.

[0075] The tractor stop condition may be the same as or different from the alarm condition. In this embodiment, the tractor stop condition is that a person in a dangerous state (e.g., a person who has fallen) is detected within a predetermined distance in the traveling direction of the asphalt finisher 100. This makes it possible to stop the travel of the asphalt finisher 100 when there is a person who is in danger of being run over by the asphalt finisher 100.

[0076] The screw control unit 108 outputs a control command for controlling the operation of the screw SC to the screw control device 55. In this embodiment, when the result of the determination by the state determination unit 103 satisfies a predetermined stop condition (hereinafter referred to as the screw stop condition), the screw control unit 108 outputs a control command for stopping the rotation of the screw SC (hereinafter referred to as the screw stop command) to the screw control device 55. This allows the asphalt finisher 100 to perform control for stopping the rotation of the screw SC when the state of a person present within the construction area of ​​the paving material satisfies the screw stop condition.

[0077] The screw stop condition may be the same as or different from the alarm condition. In this embodiment, the screw stop condition is that a person in a dangerous state (e.g., a person who has fallen) is detected within a predetermined distance ahead of the operating range of the screw SC. This makes it possible to stop the rotation of the screw SC when there is a person who may be caught in the screw SC.

[0078] <<Remote management function>> In the remote management device 300, an operation to stop the asphalt finisher 100 may be performed by a user who remotely manages the work site. In the present embodiment, the operation to stop the asphalt finisher 100 includes an operation to stop the progress of the tractor 1 or an operation to stop the rotation of the screw SC.

[0079] When an operation to stop the progress of the tractor 1 is performed by the user, the remote management device 300 transmits a control signal to stop the progress of the tractor 1 to the asphalt finisher 100. When the communication control unit 106 receives the control signal from the remote management device 300, the movement control unit 107 outputs a tractor stop command to the drive system controller 54.

[0080] When an operation to stop the rotation of the screw SC is performed by the user, the remote management device 300 transmits a control signal to stop the rotation of the screw SC to the asphalt finisher 100. When the communication control unit 106 receives the control signal from the remote management device 300, the screw control unit 108 outputs a screw stop command to the screw control device 55.

[0081] <<Log confirmation function>> The communication control unit 106 may transmit the log information stored in the log information storage unit 48a to the remote management device 300. The remote management device 300 stores the log information received from the asphalt finisher 100 and displays the past detection information and determination results on the display device according to the operation of the user who remotely manages. Thereby, the user of the remote management device 300 can confirm the past situation around the asphalt finisher 100.

[0082] The remote management device 300 displays a list of the log information received from the asphalt finisher 100. The remote management device 300 may reproduce the detection information indicating the state of a person over time. At this time, the remote management device 300 highlights the log information in which it is determined that there is a person who is determined to satisfy the reporting condition. The highlighting mode may be any mode as long as it is easy for the user to visually recognize. For example, a predetermined icon may be displayed in the time zone when a person who satisfies the reporting condition is imaged. The remote management device 300 may reproduce the detection information from the time specified by the user who remotely manages. Thereby, the user of the remote management device 300 can easily confirm the situation around the asphalt finisher 100 in the time zone when there is a person who is determined to satisfy the reporting condition.

[0083] <Detection process> The detection process executed by the asphalt finisher 100 will be described with reference to FIG. 7. FIG. 7 is a flowchart showing an example of the detection process executed by the asphalt finisher according to the embodiment.

[0084] The detection process is an example of a process in which the asphalt finisher 100 detects the construction range of the paving material and outputs information regarding the state of the detected person. The detection process is executed at a predetermined time interval. The detection process may be executed every time the imaging device 51 captures image information.

[0085] In step S1, the imaging device 51 detects the surroundings of the asphalt finisher 100. Next, the imaging device 51 outputs the detection information to the controller 50. The detection information includes the result of detecting the construction range of the paving material.

[0086] The acquisition unit 101 of the controller 50 acquires the detection information from the imaging device 51. Then, the acquisition unit 101 sends the acquired detection information to the person detection unit 102.

[0087] In step S2, the person detection unit 102 of the controller 50 receives detection information from the acquisition unit 101. Next, the person detection unit 102 detects a person existing within the construction range of the paving material based on the received detection information. Then, the person detection unit 102 sends a detection result indicating the range where the person is detected to the state determination unit 103.

[0088] In step S3, the state determination unit 103 of the controller 50 receives the detection result from the person detection unit 102. Next, the state determination unit 103 determines the state of the person detected by the person detection unit 102 based on the received detection result. The state determination unit 103 stores the log information including the detection information and the determination result in the log information storage unit 48a.

[0089] Then, the state determination unit 103 sends the detection information and the determination result to the display control unit 105 and the communication control unit 106. Also, the state determination unit 103 sends the determination result to the alarm unit 104, the movement control unit 107, and the screw control unit 108.

[0090] In step S4, the display control unit 105 of the controller 50 receives the detection information and the determination result from the state determination unit 103. Next, the display control unit 105 displays the detection information indicating the state of the person on the display device 52.

[0091] Also, the communication control unit 106 of the controller 50 receives the detection information and the determination result from the state determination unit 103. Next, the communication control unit 106 transmits the detection information and the determination result to the communication terminal 200 or the remote management device 300. The communication terminal 200 or the remote management device 300 displays the detection information indicating the state of the person on the display device based on the received detection information and determination result.

[0092] In step S5, the reporting unit 104 of the controller 50 receives the determination result from the state determination unit 103. Next, the reporting unit 104 determines whether a person in a state satisfying the reporting condition has been detected based on the received determination result. If a person in a state satisfying the reporting condition has been detected (YES), the reporting unit 104 proceeds to step S6. On the other hand, if a person in a state satisfying the reporting condition has not been detected (NO), the reporting unit 104 skips step S6 and proceeds to step S7.

[0093] In step S6, the reporting unit 104 of the controller 50 outputs a signal for reporting that a person in a state satisfying the reporting condition has been detected. For example, the reporting unit 104 outputs a signal instructing the display device 52 to display a warning indicating that there is a person determined to satisfy the reporting condition. Also, for example, the reporting unit 104 outputs a signal instructing the audio output device 56 to emit a sound or voice indicating that a person in a state satisfying the reporting condition has been detected.

[0094] In step S7, the movement control unit 107 of the controller 50 receives the determination result from the state determination unit 103. Next, the movement control unit 107 determines whether the tractor stop condition is satisfied based on the received determination result. If the tractor stop condition is satisfied (YES), the movement control unit 107 proceeds to step S8. On the other hand, if the tractor stop condition is not satisfied (NO), the movement control unit 107 skips step S8 and proceeds to step S9.

[0095] In step S8, the movement control unit 107 of the controller 50 outputs a tractor stop command to the drive system controller 54. The drive system controller 54 stops the progress of the tractor 1 according to the tractor stop command.

[0096] In step S9, the screw control unit 108 of the controller 50 receives the determination result from the state determination unit 103. Next, the screw control unit 108 determines whether or not the screw stop condition is satisfied based on the received determination result. If the screw stop condition is satisfied (YES), the screw control unit 108 proceeds to step S10. On the other hand, if the screw stop condition is not satisfied (NO), the screw control unit 108 ends the detection process.

[0097] In step S10, the screw control unit 108 of the controller 50 outputs a screw stop command to the screw control device 55. The screw control device 55 stops the rotation of the screw SC according to the screw stop command.

[0098] <Effects of the Embodiment> The asphalt finisher 100 in the present embodiment includes a tractor 1, a hopper 2 installed on the front side of the tractor 1 for receiving paving materials, a conveyor CV for conveying the paving materials in the hopper 2 to the rear side of the tractor 1, a screw SC for spreading the paving materials conveyed by the conveyor CV in the vehicle width direction on the rear side of the tractor 1, a screed 3 for leveling the paving materials spread by the screw SC on the rear side of the screw SC, and an imaging device 51 for detecting the construction range of the paving materials. The asphalt finisher 100 performs a process of outputting information regarding the state of a person detected by the imaging device 51. On one aspect, according to the present embodiment, the situation of the construction range of the paving materials can be easily known. In particular, in the construction range of the paving materials, there is a risk of danger to surrounding workers due to the operation of the asphalt finisher 100, so accidents such as workers coming into contact with the asphalt finisher 100 can be prevented.

[0099] The asphalt finisher 100 detects a person present within a predetermined distance in the traveling direction of the asphalt finisher 100, in front of the operating range of the screw SC, within a predetermined distance to the side of the screed 3, or within a predetermined distance to the side of the hopper 2. For example, if a worker is present in the traveling direction of the asphalt finisher 100, the worker may come into contact with the asphalt finisher 100. Also, for example, if a worker is present in front of the operating range of the screw SC, the worker may be caught in the screw SC. Furthermore, for example, if a worker is present on the side of the screed 3, the screed 3 may come into contact with the worker when it is extended. Also, for example, if a worker is present on the side of the hopper 2, the screed 3 may come into contact with the worker when it is opened. Therefore, according to this embodiment, it is possible to easily know that a person is present in a position where danger is likely to occur.

[0100] The state of the person includes one or more of a person not standing upright, a person who has fallen, and a person who is not looking at the asphalt finisher 100. If a person in a position where danger is likely to occur is in a position that does not allow them to take immediate evacuating action, the possibility of an abnormality such as contact occurring increases. Therefore, according to this embodiment, it is possible to easily know that there is a person in a position where danger is likely to occur and in a state where it is difficult to avoid danger.

[0101] The asphalt finisher 100 detects people using a machine-learned model that uses as training data a number of images showing people at different positions or in different states in the paving material application range. Machine learning may be able to detect people in states with similar characteristics even if they are different from a predefined state. According to this embodiment, people in dangerous states can be detected with high accuracy.

[0102] The asphalt finisher 100 notifies that a person in a state satisfying the notification conditions has been detected. According to this embodiment, a person at the work site can immediately know that there is a person in a dangerous state within the construction range of the paving material. If a person at the work site recognizes a person in a dangerous state, an accident can be avoided by, for example, alerting that person or taking rescue actions.

[0103] The asphalt finisher 100 displays information regarding the state of a person on the display device 52 of the asphalt finisher 100. According to this embodiment, the driver of the asphalt finisher 100 can immediately know that there is a person in a dangerous state within the construction range of the paving material. If the driver of the asphalt finisher 100 recognizes a person in a dangerous state, an accident can be avoided by, for example, stopping the operation of the asphalt finisher 100.

[0104] The asphalt finisher 100 transmits information regarding the state of a person to the communication terminal 200 or the remote management device 300. According to this embodiment, the user of the communication terminal 200 or the remote management device 300 can know that there is a person in a dangerous state within the construction range of the paving material at the work site or remotely. If the user of the communication terminal 200 recognizes a person in a dangerous state, an accident can be avoided by, for example, alerting that person or taking rescue actions. If the user of the remote management device 300 recognizes a person in a dangerous state, an accident can be avoided by, for example, remotely stopping the operation of the asphalt finisher 100.

[0105] The asphalt finisher 100 outputs a control command to stop the advancement of the tractor 1 or the rotation of the screw SC. According to this embodiment, if a person in a dangerous state is present within the construction range of the paving material, the advancement of the tractor 1 or the rotation of the screw SC can be stopped. For example, if a person in a dangerous state is present in the direction of advancement of the asphalt finisher 100, stopping the advancement of the tractor 1 can prevent an accident in which the person comes into contact with the asphalt finisher 100. Also, for example, if a person in a dangerous state is present ahead of the operating range of the screw SC, stopping the rotation of the screw SC can prevent an accident in which the person is caught in the screw SC.

[0106] Although the embodiment of the asphalt finisher according to the present invention has been described above, the present invention is not limited to the above embodiment. Various changes, modifications, substitutions, additions, deletions, and combinations are possible within the scope of the claims. These naturally fall within the technical scope of the present invention. [Explanation of symbols]

[0107] 1: Tractor 2: Hopper 3: Screed 48:Auxiliary storage device 50: Controller 51: Imaging device 52:Display device 53: Communication equipment 54: Drive system controller 55: Screw control device 56: Audio output device 100: Asphalt finisher 101: Acquisition Department 102: Person detection unit 103: Status determination unit 104: Reporting Department 105: Display control unit 106: Communication control unit 107: Movement control unit 108: Screw control section 200: Communication terminal 300: Remote management device CV: Conveyor SC: Screw SYS: Road Pavement System

Claims

1. A tractor, a hopper installed on the front side of the tractor for receiving paving materials, a conveyor for conveying the paving materials in the hopper to the rear side of the tractor, a screw for spreading the paving materials conveyed by the conveyor in the vehicle width direction at the rear side of the tractor on the road surface, a screed for leveling the paving materials spread by the screw at the rear side of the screw, a detection unit for detecting the construction range of the paving materials, comprising, configured to perform a process of outputting information regarding the state of a person detected by the detection unit, an asphalt finisher.

2. The detection unit is configured to detect any one or more of the persons existing within any one or more of a predetermined distance in the traveling direction of the asphalt finisher, in front of the operating range of the screw, within a predetermined distance on the side of the screed, and within a predetermined distance on the side of the hopper. The asphalt finisher according to Claim 1.

3. The detection unit is configured to detect any one or more of a person who is not standing upright, a person who has fallen, and a person whose line of sight is not directed at the asphalt finisher. The asphalt finisher according to Claim 2.

4. The person is detected by using a machine learning-trained model with a plurality of images in which the persons having different positions in the construction range of the paving materials or the persons having different states are reflected as teacher data. The asphalt finisher according to Claim 3.

5. configured to output a signal for reporting that a person whose state detected by the detection unit satisfies a predetermined condition has been detected. The asphalt finisher according to any one of Claims 1 to 4.

6. configured to display the information regarding the state on a display device of the asphalt finisher. The asphalt finisher according to any one of Claims 1 to 4.

7. configured to transmit the information regarding the state to an external device. The asphalt finisher according to any one of Claims 1 to 4.

8. configured to output a control command for stopping the travel of the tractor or the rotation of the screw. The asphalt finisher according to any one of Claims 1 to 4.

Images