Vision-based intelligent robot, working method and application

Abstract

The invention relates to the technical field of robots, in particular to a vision-based intelligent robot, a working method and application, the intelligent robot comprises a multi-joint mechanical arm body, and an end effector, an environment detection unit and a control unit which are arranged on the multi-joint mechanical arm body, and the end effector comprises a cutterhead and a high-pressure water injection unit; wherein the cutter head comprises a driving disc, a supporting block and a cutter head, a cutter face and a hammering face are arranged on the two sides of the cutter head correspondingly, the hammering face of the cutter head is matched with the high-pressure spray head to form a mud discharging pipeline capable of being sucked away by the dredging mechanism, the cutter face of the cutter head cuts up a wood structure, and the wood structure with the large size is changed into the wood structure with the small size. Therefore, sundries such as branches, aquatic plants, root systems and the like can be treated, and the device can reliably work in silt and gravel for a long time.

Description

Technical Field

[0001] This invention relates to the field of robotics, and in particular to vision-based intelligent robots, their working methods, and applications. Background Technology

[0002] China is a major agricultural country, and the water pipelines currently used in rural areas are still mainly made of concrete. These pipelines are prone to damage and siltation. After years of continuous erosion, the inner walls of these pipelines are increasingly clogged, hindering the smooth delivery of irrigation water. Surveys show that farmers now irrigate 0.067 mu (approximately 0.067 hectares) of land 2-3 hours longer than five years ago, and monthly electricity consumption has increased by 50-80 kWh. Production costs are rising annually while efficiency is declining, creating a burden on increasing yields and income. This indicates that siltation in underground pipelines significantly reduces irrigation efficiency and agricultural benefits. Severe siltation in underground irrigation pipelines creates a vicious cycle: fields far from irrigation pumping stations cannot receive sufficient irrigation water, causing pumping station motors to operate beyond their operating time or even break down, greatly increasing electricity and maintenance costs. Simultaneously, it causes severe water shortages in some fields, affecting planting time and leading to reduced crop yields or even crop death. Re-laying underground irrigation pipelines for severely silted pipelines would be prohibitively expensive. One of the more economical and effective methods is to use water pipeline inspection and dredging robots to achieve online detection, maintenance and dredging of pipelines.

[0003] Research on the silt deposits inside the pipes revealed that the main cause of siltation is the excessive amount of mud, gravel, aquatic plants, and moss in the water during agricultural water conveyance operations, leading to soil accumulation. During the dry season, animals often build nests in the pipes using leaves, branches, plant roots, and other debris. Therefore, agricultural drainage pipes are more prone to siltation and more difficult to clear than municipal pipes.

[0004] Currently, research on pipeline robots both domestically and internationally is quite in-depth, mainly focusing on structural design and motion control. For example, the research team led by Bemhard Klaassen designed a robot composed of six identical cylindrical modules with 21 degrees of freedom, named MAKRO. Compared with traditional pipeline robots, this type of pipeline robot has a unique obstacle-crossing method and walking mode. GE in the United States designed and developed a wheeled walking robot, which has a large market share in the pipeline robot market. GE named it ROVVER. This wheeled pipeline robot is available in four-wheeled and six-wheeled versions, with the four-wheeled version generally chosen for engineering applications. Due to its wheeled structure design, the robot can freely adjust its direction or turn around while working in the pipeline, allowing it to adapt to pipelines of different shapes. The large wheels also contribute to its excellent obstacle-crossing performance. The laboratory of Beijing Jiaotong University has long cooperated with Beijing Urban Drainage Group to research and develop agricultural pipeline dredging robots. It has now developed four generations of pipeline dredging robots. The fourth-generation pipeline dredging robot utilizes the principle of earthworm worm movement, using the extension and retraction of a central cylinder to propel the robot. Both the front and rear sections are equipped with support cylinders, and an auger is used to dredge the silt. The control system is electrically controlled. All of the above solutions have improved the walking method. The robot jointly developed by Changzhou University and the Municipal River and Lake Management Office has four drive wheels, a low center of gravity, and an overall length of about 800mm. This robot is a cable-driven robot, which is powered by a towed cable. The robot has a simple structure. When dredging agricultural pipelines, the front dredging cutter head dredges the pipeline. There is also a wheeled cable-threading dredging robot. This robot is driven by four wheels. The robot relies on the rotating cutter head at the front to dredge and clean the pipeline. The robot has a simple structure. Since all four wheels can provide driving force, the speed is relatively fast, reaching 5.6~5.8m / min.

[0005] In the aforementioned technical solutions, a rotating cutterhead, directly fixed to the robot body, is used to break up the silt. The drawback is that the cutterhead diameter is fixed, limiting its cleaning capability to drainage pipes of a fixed diameter. Furthermore, this cutterhead structure cannot handle agricultural water pipes containing both branches and roots, as the silt and gravel inside the pipes severely wear down the cutterhead, significantly reducing its lifespan. Moreover, the cutterhead cannot process branches, weeds, roots, or other debris. Therefore, there is an urgent need for a vision-based intelligent robot capable of handling branches, weeds, roots, and other debris, and also capable of long-term operation in silt and gravel environments to perform this specialized task. This robotic arm can be mounted on different walking mechanisms. Summary of the Invention

[0006] The purpose of this invention is to provide vision-based intelligent robots, working methods, and applications to solve the problems mentioned in the background art. The specific technical solution is as follows:

[0007] To achieve the above and other related objectives, the first objective of this invention is to provide a vision-based intelligent robot, including a multi-joint robotic arm body, and an end effector and a control unit disposed on the multi-joint robotic arm body;

[0008] The end effector includes a cutter head and a high-pressure water jet unit;

[0009] The cutter head includes a drive disc, support blocks, and a cutter head. Multiple support blocks are arranged on the circumferential surface of the drive disc. The drive disc is mounted on the main body of the multi-joint robotic arm. The angle between the rotation axis of the drive disc and the axis of the pipe is arranged at 60°-120°.

[0010] The drive disc is controlled by the control unit to perform forward or reverse rotation.

[0011] The blade head is provided with a cutting surface and a hammering surface on both sides, and the blade head is mounted on the support block so that the hammering surface and the cutting surface of the blade head face clockwise and counterclockwise directions respectively.

[0012] The high-pressure water jet unit includes a secondary arm, a squeezing block, and a high-pressure nozzle. The secondary arm is mounted on the main body of the multi-joint robotic arm and can pitch and swing on the main body of the multi-joint robotic arm. The end of the secondary arm is equipped with a squeezing block, and the squeezing block is equipped with a high-pressure nozzle that sprays downwards.

[0013] Preferably, the multi-joint robotic arm body is composed of multiple rotary joints and swing joints connected in series, used to adjust the position and attitude of the end effector in three-dimensional space; wherein,

[0014] The main body of the multi-joint robotic arm includes a rotating base, an upper arm, a middle arm, a lower arm, and a secondary lower arm;

[0015] The rotating base is provided with a large arm, and the large arm is provided with a horizontally arranged middle arm. The middle arm includes a first middle arm, a second middle arm, and a rotating shaft joint. The first middle arm and the second middle arm are connected by the rotating shaft joint. The second middle arm is provided with a pitch-swinging forearm, and the forearm is provided with an auxiliary forearm. The first middle arm is connected to the large arm, and an environmental detection unit is provided on the first middle arm. The rotating shaft joint is used to drive the rotation of the second middle arm.

[0016] Preferably, the environmental detection unit includes a camera and a searchlight, and the control unit is based on the STM32 platform and has a built-in YOLO vision system. The YOLO vision system is configured to identify the working environment by using the images captured by the camera.

[0017] Preferably, the environmental detection unit further includes a sound receiving device for monitoring ambient sound, and the control unit also has a built-in sound analysis system to collect the collision sounds generated when the cutter head is working. The sound analysis system is set to remove noise and impurities from the sound and determine the operating status of the cutter head based on the sound.

[0018] Preferably, the hammering surface is further provided with multiple protrusions.

[0019] Preferably, the forearm is also provided with a mudguard.

[0020] Preferably, the forearm is an electrically telescopic structure.

[0021] Preferably, the cutter head is detachably mounted on the support block.

[0022] A second objective of this invention is to provide a method for operating a vision-based intelligent robot as described in any of the preceding claims, comprising the following steps:

[0023] S1: Control the multi-joint robotic arm to move the end effector to the location of the pipe blockage;

[0024] S2: Control the cutter head to rotate forward or backward based on the type of blockage identified by the environmental detection unit;

[0025] S3: If the blockage is mud, sand, gravel, or hardened lumps, control the cutter head to reverse and use the hammer surface to break it up.

[0026] S4: If the blockage is made of wood, control the high-pressure nozzle to flush out the wood and control the cutter head to rotate forward, using the cutting blade to cut;

[0027] S5: Controls the high-pressure nozzle to spray high-pressure water to break up the blockage into slurry, which is then easily sucked out by the dredging mechanism;

[0028] S6: Real-time monitoring of the workload and ambient noise of the cutter head, adjusting the movement speed of the robotic arm body and the rotation speed of the cutter head according to changes in load and noise.

[0029] A third objective of this invention is to protect the application of a vision-based intelligent robot as described in any of the preceding claims in agricultural drainage pipes.

[0030] The vision-based intelligent robot provided by this invention has the following beneficial effects:

[0031] 1. Based on vision and a multi-degree-of-freedom robotic arm, this invention achieves precise positioning and adaptive operation in complex environments, significantly improving the robotic arm's environmental adaptability. Furthermore, by combining vision and sound detection, it enables intelligent decision-making and real-time adjustment during the robotic arm's operation, reducing the need for manual intervention.

[0032] 2. This invention utilizes the hammering surface of the cutter head to break up the slabs and clumps at the blockage location. Multiple protrusions are set on the hammering surface to further enhance the breaking effect, thus dispersing the sand and gravel. The middle arm drives the cutter disc to move along the circumference of the pipe, cleaning the inner wall of the pipe. At the same time, the high-pressure nozzle sprays high-pressure water to disperse the slabs and clumps, forming mud that can be sucked up by the dredging mechanism and discharged from the pipe. The gravel broken up by the cutter disc is more evenly dispersed in the silt and can be smoothly sucked up by the dredging mechanism without easily getting stuck. Using the YOLO vision system and sound analysis system, wooden structures can be detected. The cutting surface of the cutter head is used to cut the wooden structures, turning large wooden structures into smaller pieces that can be discharged by the dredging mechanism. This invention can handle debris such as branches, weeds, and roots, and can work reliably in mud, sand, and gravel for a long time. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0034] Figure 1 This is a schematic diagram of the structure of the present invention;

[0035] Figure 2 This is a schematic diagram of the cutter head described in this invention. Detailed Implementation

[0036] The vision-based intelligent robot proposed in this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of this invention will become clearer from the following description. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of this invention.

[0037] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0038] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the state, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0039] The core idea of ​​this invention is to construct an intelligent robot using a multi-axis robotic arm and an environmental detection unit as the main components, and a high-speed rotating cutter head as the end effector. The cutter head is specially designed to act as a cutter when the cutter head rotates forward, using the sharp edge of the cutter to cut and crush the wooden structure. When the cutter head rotates in reverse, it acts as a breaker hammer, breaking up the hardened silt. The strength of the breaker hammer is used to resist the abrasion of mud and gravel. High-pressure water jets are used to reveal the wooden structure in the silt. A visual recognition system is used to identify the wooden structure. A controller controls the forward and reverse rotation of the cutter head.

[0040] like Figure 1-2 As shown, the vision-based intelligent robot includes a multi-joint robotic arm body 1, an end effector 2, an environmental detection unit 3, and a control unit mounted on the multi-joint robotic arm body 1. The multi-joint robotic arm body 1 can be connected to a walking mechanism. The end effector 2 includes a cutter head 21 and a high-pressure water jet unit 22. The cutter head 21 includes a drive disc 211, support blocks 212, and a cutter head 213. Multiple support blocks 212 are arranged on the circumferential surface of the drive disc 211. The drive disc 211 is mounted on the multi-joint robotic arm body 1. The angle between the rotation axis of the drive disc 211 and the axis of the pipe is arranged at 60°-120°. The drive disc 211 is controlled by the control unit to perform forward or reverse rotation. The cutter head 213 has a cutting surface 213a and a hammering surface on both sides. The hammering surface 213b of the blade 213 is detachably mounted on the support block 212, so that the hammering surface 213b of the blade 213 faces counterclockwise and the cutting surface 213a of the blade 213 faces clockwise. When the drive disc 211 rotates in reverse, the hammering surface 213b of the blade 213 impacts the hardened silt. When the drive disc 211 rotates clockwise, the cutting surface 213a of the blade 213 cuts the wooden structure. The high-pressure water jet unit 22 includes a secondary arm 221, a pressing block 222, and a high-pressure nozzle 223. The secondary arm 221 is mounted on the multi-joint robotic arm body 1 and can swing up and down on the multi-joint robotic arm body 1. The end of the secondary arm 221 is provided with a pressing block 222, and the pressing block 222 is provided with a downward-spraying high-pressure nozzle 223.

[0041] In some embodiments, the multi-joint robotic arm body 1 is composed of multiple rotary joints and swing joints connected in series, used to adjust the position and attitude of the end effector 2 in three-dimensional space; the multi-joint robotic arm body 1 includes a rotating base 11, a large arm 12 is provided on the rotating base 11, a horizontally arranged middle arm 13 is provided on the large arm 12, the middle arm 13 includes a first middle arm 131, a second middle arm 132, and a rotary axis joint 133, the first middle arm 131 and the second middle arm 132 are connected by the rotary axis joint 133, the second middle arm 132 is provided with a pitch swinging forearm 14, the forearm 14 is provided with a secondary forearm 221, the first middle arm 131 is connected to the large arm 12, and an environmental detection unit 3 is provided on the first middle arm 131, the rotary axis joint 133 is used to drive the second middle arm 132 to rotate 360°.

[0042] In some embodiments, the environmental detection unit 3 includes a camera 31 and a searchlight 32. The control unit is based on the STM32 platform and has a built-in YOLO vision system. The YOLO vision system is configured to use the image captured by the camera 31 to identify various blockages and locate the leakage location, providing auxiliary control for the operator. Since gravel, branches, etc., have been soaked in silt for too long and their colors tend to be uniform, the YOLO vision system is prone to missing identification. Therefore, in some embodiments, the environmental detection unit 3 also includes a sound receiving device 33 for monitoring ambient sound. The control unit also has a built-in sound analysis system to collect the sound generated when the cutter head 21 collides with the blockage during operation. The sound analysis system is configured to remove impurities and reduce noise from the sound, and determine whether the cutter head 21 is in contact with silt, wooden structures, gravel, etc., based on the sound, and remind the operator. In order to prevent the cutter head 21 from splashing silt onto the environmental detection unit 3, a mudguard 141 is also provided on the forearm 14.

[0043] It should be noted that when this invention is applied to dredging agricultural drainage pipes, it needs to work in conjunction with a traveling mechanism, a sludge removal mechanism, and a water pump mechanism. During operation, the rotating base 11 is first installed on the traveling mechanism, which can be an existing four-wheeled, six-wheeled, or tracked pipe trolley. The water pump mechanism is connected to a high-pressure nozzle 223 via a high-pressure pipeline. The sludge removal mechanism can be arranged on the traveling mechanism and uses a suction pump, such as a spiral conveying pump or a negative pressure mud pump. When it reaches the blockage location… The upper boom 12 moves the middle boom 13 to the center of the pipe, and the lower boom 14 swings downward so that the lower edge of the cutter head 21 is below the height of the blockage. The traveling mechanism moves forward, and the cutter head 21 rotates counterclockwise, using the hammering surface 213b of the cutter head 213 to break up the slabs at the blockage location. To further improve the breaking effect, multiple protrusions 213c are provided on the hammering surface 213b to disperse the sand and gravel. The rotating shaft joint 133 drives the second middle boom 132 to rotate, so that the cutter head 21 moves along the circle of the pipe. The device moves in a circumferential direction to clean the inner wall of the pipe. Simultaneously, the high-pressure nozzle 223 sprays high-pressure water to disperse agglomerated clumps, forming sludge that can be sucked up by the sludge removal mechanism and discharged from the pipe. For gravel deposited at the bottom of the pipe, without the dispersing action of the cutter head 21, the accumulated gravel is easily stuck in the sludge removal mechanism. However, the gravel dispersed more evenly in the sludge by the cutter head 21 can be smoothly sucked up by the sludge removal mechanism and is less likely to get stuck. When the YOLO vision system or sound analysis system detects... When the wooden structure is in use, the auxiliary arm 221 is moved down, and the pressing block 222 is used to press down the wooden structure. The high-pressure nozzle 223 washes the wooden structure with high pressure, cleaning the surrounding mud, sand and gravel. The cutter head 21 is controlled to rotate clockwise, and the cutting surface 213a of the cutter head 213 is used to cut the wooden structure. The arm 14 adopts an electric telescopic structure. The arm 14 gradually extends, pushing the cutter head 21 forward to cut the wooden structure, turning these large wooden structures into small pieces of wooden structure, which can be discharged through the dredging mechanism.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vision-based intelligent robot, characterized in that, It includes a multi-joint robotic arm body, and an end effector and control unit mounted on the multi-joint robotic arm body; The end effector includes a cutter head and a high-pressure water jet unit; The cutter head includes a drive disc, support blocks, and a cutter head. Multiple support blocks are arranged on the circumferential surface of the drive disc. The drive disc is mounted on the main body of the multi-joint robotic arm. The angle between the rotation axis of the drive disc and the axis of the pipe is arranged at 60°-120°. The drive disc is controlled by the control unit to perform forward or reverse rotation. The blade head is provided with a cutting surface and a hammering surface on both sides, and the blade head is mounted on the support block so that the hammering surface and the cutting surface of the blade head face clockwise and counterclockwise directions respectively. The high-pressure water jet unit includes a secondary arm, a squeezing block, and a high-pressure nozzle. The secondary arm is mounted on the main body of the multi-joint robotic arm and can pitch and swing on the main body of the multi-joint robotic arm. The end of the secondary arm is equipped with a squeezing block, and the squeezing block is equipped with a high-pressure nozzle that sprays downwards.

2. The vision-based intelligent robot according to claim 1, characterized in that, The multi-joint robotic arm is composed of multiple rotary joints and swing joints connected in series, used to adjust the position and orientation of the end effector in three-dimensional space; wherein, The main body of the multi-joint robotic arm includes a rotating base, an upper arm, a middle arm, a lower arm, and a secondary lower arm; The rotating base is provided with a large arm, and the large arm is provided with a horizontally arranged middle arm. The middle arm includes a first middle arm, a second middle arm, and a rotating shaft joint. The first middle arm and the second middle arm are connected by the rotating shaft joint. The second middle arm is provided with a pitch-swinging forearm, and the forearm is provided with an auxiliary forearm. The first middle arm is connected to the large arm, and an environmental detection unit is provided on the first middle arm. The rotating shaft joint is used to drive the rotation of the second middle arm.

3. The vision-based intelligent robot according to claim 2, characterized in that, The environmental detection unit includes a camera and a searchlight. The control unit is based on the STM32 platform and has a built-in YOLO vision system. The YOLO vision system is configured to identify the working environment by using images captured by the camera.

4. The vision-based intelligent robot according to claim 3, characterized in that, The environmental detection unit also includes a sound receiving device for monitoring ambient sound. The control unit also has a built-in sound analysis system to collect the collision sounds generated when the cutter head is working. The sound analysis system is set to remove noise and impurities from the sound and determine the operating status of the cutter head based on the sound.

5. The vision-based intelligent robot according to claim 1, characterized in that, The hammering surface is also provided with multiple protrusions.

6. The vision-based intelligent robot according to claim 1, characterized in that, The forearm is also equipped with a mudguard.

7. The vision-based intelligent robot according to claim 1, characterized in that, The forearm is an electrically telescopic structure.

8. The vision-based intelligent robot according to claim 1, characterized in that, The cutter head is detachably mounted on the support block.

9. The working method of the vision-based intelligent robot according to claims 1-8, characterized in that, Includes the following steps: S1: Control the multi-joint robotic arm to move the end effector to the location of the pipe blockage; S2: Control the cutter head to rotate forward or backward based on the type of blockage identified by the environmental detection unit; S3: If the blockage is mud, sand, gravel, or hardened lumps, control the cutter head to reverse and use the hammer surface to break it up. S4: If the blockage is made of wood, control the high-pressure nozzle to flush out the wood and control the cutter head to rotate forward, using the cutting blade to cut; S5: Controls the high-pressure nozzle to spray high-pressure water to break up the blockage into slurry, which is then easily sucked out by the dredging mechanism; S6: Real-time monitoring of the workload and ambient noise of the cutter head, adjusting the movement speed of the robotic arm body and the rotation speed of the cutter head according to changes in load and noise.

10. An application of a vision-based intelligent robot as described in any one of claims 1-8 in agricultural drainage pipes.

Images