379 results about "Obstacle crossing" patented technology

The invention discloses a catastrophe rescue robot digital twin control method and system, and belongs to the technical field of robot intelligent control. Comprising a virtual-real mapping module which processes the body state of a robot and catastrophe scene point cloud data in real time, and constructs and synchronously updates a digital twinborn body containing a terrain elevation map at a cloud end; the hierarchical decision-making module is used for making a decision in the digital twinborn body by utilizing a hierarchical reinforcement learning model, and outputting a wheel-leg movement mode and an excavation arm cooperation instruction by a high-level strategy network according to the terrain complexity; and the safety deduction module introduces a safety reinforcement learning mechanism, deduces and verifies the generated action instruction in the virtual environment, and restrains the dangerous action by controlling a barrier function. According to the method, the decision risk of the heavy robot in the unstructured catastrophic terrain can be reduced, the stability of motion control is improved, and a high-safety control strategy is provided for various complex obstacle crossing actions based on simulation prediction.

The invention discloses a magnetic adsorption robot and a child-mother robot for bridge detection, and belongs to the technical field of bridge detection, the magnetic adsorption robot comprises two vehicle bodies arranged front and back in the advancing direction of the magnetic adsorption robot, and a connecting joint with a self-adaptive angle is connected between the two vehicle bodies; mounting parts are arranged at the bottoms of the vehicle bodies, wheel carriers are hinged to the two sides of each mounting part through hinges, and each wheel carrier is provided with a magnetic adsorption wheel; a detection device, a control device and a battery module are arranged on the vehicle body, the control device is used for controlling the magnetic adsorption robot to move on the bridge, the detection device is used for detecting the bridge, and the battery module is used for providing power support. By arranging the connecting joints and the hinges, the magnetic adsorption wheels can be tightly attached to the adsorption face all the time, the adsorption capacity and the obstacle crossing capacity are improved, and the reliable inspection capacity is achieved; and meanwhile, the two magnetic adsorption robots with different sizes are arranged to form the child-mother robot, so that the safety and the cruising ability are further enhanced.

The invention relates to the technical field of agricultural machinery, in particular to a rotary gravity center regulating crawler platform for hills and mountains, which sequentially comprises a crawler chassis assembly, a gravity center regulating assembly and a functional module mounting platform from bottom to top, the gravity center adjusting layer is installed on the chassis rack and located in the space between the crawler belts on the two sides, the gravity center adjusting layer is composed of a supporting base, a slewing bearing module, a linear lead screw module, a battery balancing weight and a machine shell, and the balancing weight is preferably a power battery pack; and the functional module mounting platform is fixed at the upper end of the gravity center adjusting assembly. By adjusting the angular and radial distribution of the counterweights, the center-of-gravity position of the whole machine under the working conditions of cross slopes, longitudinal slopes and obstacle crossing is changed, the stability and driving safety of the crawler belt platform in complex terrains of hills and mountains are improved, and the crawler belt platform can be provided with different functional modules according to operation objects. The method is suitable for various hilly and mountainous region agriculture and engineering application scenes.

The utility model provides cleaning equipment. The cleaning equipment comprises a main machine body, driving wheels and obstacle crossing wheel devices. The driving wheels are arranged on the main machine body and used for driving the main machine body to move; the obstacle crossing wheel device comprises a driving mechanism and an obstacle crossing wheel, the driving mechanism is used for driving the obstacle crossing wheel to rotate, the obstacle crossing wheel is located in front of the gravity center of the cleaning equipment, the obstacle crossing wheel comprises an obstacle crossing state and an avoiding state, and in the obstacle crossing state, the obstacle crossing wheel rotates to be supported on the ground; and in the avoiding state, the obstacle crossing wheels are separated from the ground. The obstacle crossing of the cleaning equipment can be realized under the pushing of the driving wheel only by driving the obstacle crossing wheel through the driving mechanism, the number of types of parts for realizing the obstacle crossing function is small, the structure is simple, and the operation stability of the cleaning equipment is favorably improved.

The present disclosure belongs to the technical field of intelligent cleaning equipment, and in particular to a swimming pool cleaning device of an underwater robot. The present disclosure facilitates maintenance of the filter screen and the impellers through the detachable structure of the upper cover and the lower cover, and forms linkage with the inductor brackets and the through holes, and automatically separates the inductor brackets during disassembly and assembly, which simplifies assembly and ensures connection stability. The guide wheel system significantly improves the travel stability and obstacle crossing ability of the device on the bottom of the pool through the precise cooperation of the wheel grooves and the caster round shafts. The dual-motor drive cooperates with the concentrical drainage enclosures to form an efficient vortex, which ensures that the cleaning ability can still be maintained when a single motor fails, and greatly improves the efficiency of discharging fine impurities.

The application aims to provide a single-motor driven multi-mode ornithopter simulating the flight posture of insects and birds, comprising a fuselage, a foldable wing mechanism, a flapping wing and ground walking mechanism, and a steering tail wing mechanism; the flapping wing and ground walking mechanism is a spatial five-link mechanism; in the flight state, the motor outputs power, the five-link mechanism drives the wing to complete the flapping action; in the ground walking state, the electromagnet connection hole is not electrified, the connecting rod driving wheel and the electromagnet right-angle connecting shaft are disconnected, the wing is in the folded state, and the rotation output of the motor is the rotation movement of the driving connecting rod driving wheel; the application integrates the connecting rod and the wheel, realizes the function of single-motor driving two modes, simplifies the mechanism complexity, reduces the manufacturing difficulty, lightens the ornithopter weight, and has good obstacle crossing ability and can cross narrow space.

The invention provides an obstacle crossing mechanism, a driving wheel module and cleaning equipment. The obstacle crossing mechanism comprises a mounting arm, a driving part, a lower supporting part and a rotating part. The driving part is rotationally arranged on the mounting arm. The rotating piece is rotationally connected with the mounting arm to form a first rotating axis. The lower supporting piece is rotationally connected with the driving piece to form a second rotating axis, and the second rotating axis and the first rotating axis are arranged in a spaced mode. The rotating piece is rotationally connected with the mounting arm, and the lower supporting piece is further movably connected with the rotating piece in the axial direction of the lower supporting piece. The driving piece rotates during obstacle crossing, so that the lower supporting piece swings backwards and abuts against the walking face downwards. When the walking wheels make contact with an obstacle, the obstacle crossing mechanism obliquely presses the walking face downwards, the walking face generates obliquely-upward counter-acting force, the counter-acting force lifts the walking wheels on one hand and drives the walking wheels to move forwards on the other hand, and the driving force in the advancing direction is increased; due to the fact that the height of the walking wheels is increased and the driving force in the advancing direction is increased, the obstacle crossing performance of the driving wheel module is improved.

The invention discloses a wheel-legged cargo carrying robot. The robot comprises a robot body module, a thigh module, a shank module and a transmission mechanism. The thigh module comprises a thigh structure which is movably connected with the web plate. The shank module comprises a shank structure, and the shank structure comprises a shank skeleton, a hub motor, wheels and a damping suspension. And the lower leg skeleton is provided with a limiting and transmission bulge and is movably connected with the lower end of the thigh structure. The wheels are driven by hub motors and installed at the lower ends of the shank skeletons through damping hangers. The transmission mechanism comprises a first driving piece, a second driving piece, a crank rocker assembly and a connecting rod. The robot has a wheel type motion mode, a leg type motion mode and a wheel-leg composite motion mode, and mode switching and obstacle crossing walking are achieved by controlling the first driving part, the second driving part and the hub motor to act cooperatively. The modular and lightweight design is adopted, the terrain trafficability is remarkably enhanced, motion control is accurate and reliable, the structure is compact, and assembly and maintenance are convenient.

The invention relates to an AUV (Autonomous Underwater Vehicle) crawling system and application thereof. The plurality of translation mechanisms are mounted on the machine body; in the plurality of lifting mechanisms, the plurality of lifting mechanisms are respectively connected with the movable ends of the plurality of translation mechanisms and the machine body; the multiple wheel set opening and closing mechanisms are connected with the fixed ends of the multiple lifting mechanisms correspondingly. The plurality of walking wheel sets are respectively connected with the plurality of wheel set opening and closing mechanisms; the multiple auxiliary clamping mechanisms are connected with the multiple wheel set opening and closing mechanisms correspondingly. In the obstacle-avoiding crawling process of each obstacle-crossing unit on the pipe / cable, the walking wheel set and the auxiliary clamping mechanism are driven to swing in sequence, the height position of the wheel set opening and closing mechanism relative to an obstacle is adjusted through the lifting mechanism, and the lifting mechanism is driven to cross the obstacle through the translation mechanism. And the obstacle crossing units repeat the actions, so that the AUV crawling system completes inchworm type obstacle crossing on the pipe / cable. The AUV crawling system can adapt to crawling of pipes / cables in different environments, is high in obstacle crossing capacity and belongs to the technical field of pipe / cable inspection.

The wall-climbing robot comprises a robot main machine, a vertical obstacle crossing platform, a plurality of climbing obstacle crossing units, a control unit and an operation executing mechanism, the vertical obstacle crossing platform is fixedly connected to one side of the robot main machine, and the vertical obstacle crossing platform is perpendicular to the robot main machine. The multiple climbing obstacle crossing units are sequentially arranged at intervals in the length direction of the vertical obstacle crossing platform, all the climbing obstacle crossing units are in sliding fit with the vertical obstacle crossing platform, the climbing obstacle crossing units are arranged on the robot host through the vertical obstacle crossing platform, and the control unit is arranged in the robot host. The control unit is used for controlling the climbing obstacle crossing units to act alternately. The climbing obstacle-crossing unit is in sliding fit with the vertical obstacle-crossing platform, so that stable movement and obstacle-crossing operation of the robot host along the wall surface are realized, the path planning of the operation executing mechanism is greatly optimized, and the robot is particularly suitable for carrying a heavy operation executing mechanism.

The invention relates to the technical field of robots, in particular to a safety inspection robot system which comprises a circumferential self-adaption mechanism, at least three radial self-adaption mechanisms and a guide walking mechanism, the circumferential self-adaption mechanism comprises at least three first telescopic rods which are sequentially hinged end to end, and the radial self-adaption mechanisms are connected with the guide walking mechanism. Each group of radial self-adaptive mechanism comprises at least two second telescopic rods which are matched for use, each second telescopic rod comprises a fixed cylinder, an actuating rod, an elastic element and a flexible rope, and the guide walking mechanism comprises a walking crawler belt, a walking wheel and a connecting rod structure; the circumferential self-adaptive mechanism surrounds the pipeline, the position of the actuating rod is adjusted, the elastic restoring force of the reset spring is combined, it is guaranteed that the radial self-adaptive mechanism is attached to the surface of the pipeline with constant and reliable pressing force all the time, and the stability of movement and detection is guaranteed; meanwhile, the walking wheels are driven by the connecting rod structures to rotate to be attached to the surface of the pipeline for obstacle crossing, and therefore continuous and comprehensive detection can be conducted.

The utility model relates to the technical field of crown block moving maintenance devices, and discloses a crown block track monorail clamping obstacle crossing moving chassis which comprises a chassis body, a clamping mechanism is arranged on the chassis body and comprises a clamping motor, a clamping gear and two sets of clamping racks, the clamping gear is meshed with the two sets of clamping racks at the same time, and the clamping motor is arranged on the chassis body. Clamping wheels are rotationally arranged on the two groups of clamping racks; a driving system is further arranged on the chassis and comprises a driving motor, the driving motor is connected with a driving gear, a driving connecting rod is rotationally arranged on the chassis, a transmission gear is fixedly arranged on the driving connecting rod, and the transmission gear is meshed with the driving gear; the chassis is further provided with a driving mechanism, the driving mechanism comprises a first gear and a driving wheel, the driving wheel is fixedly connected with a second gear, and the second gear is meshed with the first gear. According to the utility model, the problem that the inspection and maintenance work of the crown block track is influenced because the movable chassis is blocked by a pressing plate bolt or an irregular joint and cannot normally move on the track can be solved.

This invention discloses an obstacle-crossing robot for use in confined spaces inside pipelines, comprising a robot body, a drive system, a navigation and positioning system, and a control system. The drive system includes a main walking mechanism and a front swing arm mechanism. The front swing arm mechanism is coupled to the main walking mechanism via a differential mechanism, enabling it to change its posture relative to the robot body and possessing independently driven swing arm tracks. The control system is configured to acquire real-time posture information during the posture change of the front swing arm mechanism and dynamically adjust the parameters of the sensors in the navigation and positioning system based on this real-time posture information to compensate for the impact of sensor pose changes caused by swing arm movements on positioning accuracy. This invention solves the technical problem of decreased positioning accuracy due to swing arm movements in existing technologies through the coordinated control of swing arm obstacle crossing and navigation and positioning, improving the positioning accuracy and obstacle-crossing stability of the robot during pipeline inspection, and is suitable for internal inspection and testing tasks of various pipelines.

The application discloses a robot chassis and a robot, and the robot chassis comprises a vehicle frame, swing arms arranged on opposite sides of the vehicle frame, different-speed rotation between the swing arms, rudders connected with the swing arms, synchronous rotation between the rudders, hub motors connected with the swing arms through the rudders, rotation of the hub motors according to a preset speed ratio, and a connecting rod mechanism arranged on the vehicle frame, connected with the swing arms, and matched with the swing arms to enable the different-speed rotation between the swing arms. The hub motors are connected with the swing arms through the rudders, the rudders of front and rear wheels are synchronously rotated, the front and rear wheels are deflected by a certain angle, the hub motors can rotate according to a certain speed ratio, and the whole machine is driven to turn during movement and turn in place. Meanwhile, the connecting rod mechanism is matched with the different-speed rotation between left and right swing arms, the pitch angle of the chassis is at the middle value of the inclination angles of the left and right swing arms, and the obstacle crossing ability and the anti-rollover ability of the chassis are effectively improved.

The application discloses a kind of based on embodiment intelligence's self-adapting rocker arm type carrying robot control method and system, belong to robot control field, the method includes: by being carried on robot car body and rocker arm mechanism on multi-modal sensor array obtains first perception dataset and second state dataset, extracts historical configuration matrix and exports feedforward configuration instruction to realize feedforward pre-adjustment;During obstacle crossing and load period, generate and execute the non-symmetrical compensation instruction of driving two sides rocker arm to carry out asymmetric telescoping and lifting;In distress period, according to the attribute of physical root node, match the corresponding quasi-physical escape sequence from state transition matrix and execute escape action;In edge control node, running reinforcement learning network module to output original speed instruction vector.The application guarantees the control completeness when facing new terrain, improves the accuracy of compensation action.

The utility model relates to the field of curtain wall cleaning, in particular to an obstacle crossing device for glass curtain wall cleaning equipment. Comprising a plurality of first elastic wheels arranged on the top face of an equipment outer frame, a first adjusting assembly arranged on the equipment outer frame and used for driving the first elastic wheels to move towards the glass curtain wall, and a plurality of second elastic wheels arranged on the bottom face of the equipment outer frame. The second adjusting assembly is arranged on the equipment outer frame and used for driving the second elastic wheel to move towards the glass curtain wall, the supporting wheels are arranged on the side, facing the glass curtain wall, of the equipment outer frame, and an anti-collision wheel located on the side edge of the equipment outer frame is arranged between the first elastic wheel and the supporting wheels. The cleaning equipment has the effects that the practicability and adaptability of the cleaning equipment are effectively improved, the possibility that the glass curtain wall is damaged due to collision during obstacle crossing of the equipment is reduced, and the operation safety of the equipment is improved.

The invention discloses a wheel diameter adjustable trolley with sensing-deformation composite spokes and a control method. Each wheel of the trolley is composed of a wheel axle and a plurality of wheel spokes distributed in a circumferential array mode, and the wheel spokes are of a three-layer composite structure: a driving layer composed of a liquid crystal elastomer is responsible for generating deformation under thermal stimulation; the rigidity-variable layer formed by a rigidity-adjustable polymer realizes hardening locking through ultraviolet irradiation, and realizes reversible adjustment of rigidity through heating; the sensing layer is formed by a multi-walled carbon nanotube grid, can be integrally formed with the driving layer and is used for monitoring the wheel diameter change in real time; according to the control method, the trolley can be switched between the high obstacle crossing ability of the large wheel form and the high load stability of the small wheel form. According to the method, the motion performance under the complex terrain is remarkably improved, and the method has wide application prospects in the field of adaptive mobile robots.

The application discloses a wheel-leg type air-ground integrated reconnaissance robot based on super-spiral sliding mode. The multi-draught type flight frame adopts a four-rotor structure, and the wheel-leg type chassis is fixed below the multi-draught type flight frame and adopts a wheel-leg type structure connected through connecting rods. The multi-draught type flight frame is used for realizing the flight function of the robot, transmitting high-definition image information in real time, and assisting the robot in obstacle crossing; the wheel-leg type chassis realizes the change of the height of the robot through the control of the steering engine on the connecting rods, so as to provide multi-level visual angles for ground reconnaissance; the wheel-leg type air-ground integrated reconnaissance robot can freely switch between the flight mode and the ground mode, the ground mode is the main action mode of the robot, and the flight mode is used for assisting in obstacle crossing and flight entering, so that the endurance of the robot is greatly improved; the self-balancing control of the wheel-leg type chassis adopts a method based on super-spiral sliding mode control, has strong robustness, can be used for various reconnaissance tasks, and makes up for the shortage of the existing robot system.

The utility model discloses a photovoltaic robot walking chassis, including power element, provides first rotation drive force, and the rotation output shaft of first rotation drive force points to ground direction, power element is connected with drive steering element, and first rotation drive force changes as the second rotation drive force of rotation output shaft parallel ground direction through drive steering element, drive steering element is connected with differential transmission assembly, and differential transmission assembly changes as the third rotation drive force of rotation output shaft vertical to first rotation drive force and second rotation drive force with second rotation drive force, and third rotation drive force drives the rotation of output shaft rotation connected in differential transmission assembly, the utility model drives the power through multiple commutation, makes power element compact arrangement of angle adjustment, reduces the device volume, and has the walking wheel of differential rotation, when encountering irregular protruding object, does not appear single wheel suspension and leads to drive power loss, and improves the obstacle crossing ability and whole structural reliability.

The invention relates to the technical field of magnetic memory detection devices, in particular to a self-adaptive variable-diameter rotatable pipeline internal magnetic memory detection robot which comprises a magnetic memory detection mechanism connected with a power mechanism through a steering mechanism. The power mechanism comprises a second mounting seat, and a plurality of walking assemblies are arranged on the peripheral side of the second mounting seat; the steering mechanism comprises a left end connecting piece connected with the first installation base and a right end connecting piece connected with the second installation base, and the left end connecting piece is hinged to the right end connecting piece. According to the invention, the magnetic memory detection mechanism can realize self-adaptive variable diameter and omnibearing rotation detection functions at the same time during working; the steering mechanism is connected with the magnetic memory detection mechanism and the power mechanism, so that the adaptability and trafficability of the whole robot at different bent pipes can be realized; the power mechanism can achieve free lifting and obstacle crossing functions of the crawler belt and overall power of the robot.

The utility model discloses a multi-purpose unmanned vehicle, and relates to the technical field of unmanned vehicle tools, and the unmanned vehicle comprises a vehicle frame mechanism which comprises a chassis structure; the two walking mechanisms are connected with the bottom of the chassis structure, each walking mechanism comprises a driving assembly and a supporting assembly which are sequentially arranged in the advancing direction of the unmanned vehicle, and the driving assemblies are located on the rear portion of the unmanned vehicle relative to the supporting assemblies. An auxiliary supporting assembly connected with the chassis structure is arranged between the driving assembly and the supporting assembly, and the driving assembly is in transmission connection with the supporting assembly through a crawler belt. The chassis structure is located in the area where the supporting assemblies and the auxiliary supporting assemblies are located, the driving assembly is located outside the chassis structure, and when the unmanned vehicle climbs or crosses over obstacles, the gravity center of the unmanned vehicle is located in the area formed by the supporting assemblies and the auxiliary supporting assemblies of the two walking mechanisms; according to the design form, the unmanned vehicle has excellent obstacle crossing and climbing performance.

The utility model relates to the technical field of fireproof separators, in particular to a comprehensive pipe gallery fireproof separator which comprises a door frame, an adjusting assembly is fixedly arranged on one side of the door frame, a fireproof door body is hinged to one side of the adjusting assembly, the adjusting assembly comprises two installation bases, and positioning sleeves are fixedly connected to the front end faces of the installation bases. A rotating shaft is rotatably connected between the two positioning sleeves and comprises a mounting shaft, an upper adjusting shaft is fixedly connected to the top end of the mounting shaft, a lower adjusting shaft is fixedly connected to the bottom end of the mounting shaft, a connecting seat is fixedly connected to the middle of the mounting shaft, an obstacle crossing assembly is fixedly connected to the bottom of the mounting shaft, and a supporting column is arranged below the obstacle crossing assembly; according to the fireproof door, the problem that the fireproof door body cannot be normally opened due to the fact that the fireproof door body is clamped by falling small objects or other obstacles is effectively avoided, and potential safety hazards are reduced.

This application provides an obstacle-crossing control method for cleaning equipment and a cleaning equipment itself. The method includes: when a multi-tiered obstacle is identified and meets the obstacle-crossing conditions, controlling the cleaning equipment to move towards the multi-tiered obstacle; for each tier of the multi-tiered obstacle, after moving to a point where at least a portion of the cleaning equipment engages with the current tier of the multi-tiered obstacle, and after the obstacle-crossing wheels engage with the current tier, controlling the drive wheels to move forward, so that the cleaning equipment crosses the current tier and at least a portion of the cleaning equipment engages with the next tier of the multi-tiered obstacle; in response to the cleaning equipment crossing the current tier and at least a portion of the cleaning equipment engaging with the next tier, updating the next tier to the new current tier, and repeating the steps of engaging the obstacle-crossing wheels and moving the drive wheels until the cleaning equipment crosses all tiers of the multi-tiered obstacle. This method is used to achieve stable and efficient crossing of obstacles with heights exceeding conventional obstacle-crossing limits, improving cleaning coverage and equipment operational reliability.

The application discloses a power transmission line inspection machine control method and system, and relates to the technical field of power transmission line operation and maintenance, including the following steps: acquiring visual and conductor internal state sensing data; positioning obstacles on the travel path based on the visual sensing data, and identifying conductor defects based on the conductor internal state sensing data; the identification of conductor defects includes collecting real-time motion state signals and real-time eddy current signals; the real-time eddy current signals are dynamically compensated, and the purified eddy current signals are divided into multiple signal base elements; the base element feature vectors of the signal base elements are extracted for dynamic matching, candidate defect types are screened, and final judgment is made according to the base element feature vector change mode; and integrated obstacle crossing and detection collaborative control instructions are generated and executed according to the obstacle recognition result. The application effectively solves the problem that the detection function of the inspection machine is interrupted during the obstacle crossing process, a data blind area is formed, and the internal health condition of the conductor cannot be completely evaluated.

The present application relates to the technical field of automatic control and machine vision, and particularly relates to a path planning and blind area elimination method for a crawler-type cleaning robot, which comprises the following steps: collecting a photovoltaic module surface image through a vision sensor arranged on a robot body, extracting photovoltaic cell grid line features and establishing a local coordinate system; generating a path reference line based on the grid line direction, and calculating a lateral deviation and an angle deviation between a current position of the robot and the path reference line in real time; dynamically adjusting a robot walking path through a fuzzy control method; dynamically adjusting a cleaning interval by calculating a current cleaning width and an expected overlap rate; and executing an obstacle crossing or board crossing decision based on an inclination angle data when sensor data conflicts. A multi-sensor data fusion technology improves the anti-interference capability of the robot, so that the robot can stably operate in a strong light, a reflective light and the like.

The wheel-foot conversion multi-legged robot comprises a body, the side surface of the body is provided with a plurality of driving parts distributed in the circumferential direction, a multi-joint leg, the proximal end of the multi-joint leg is hinged to the driving part, the multi-joint leg comprises a plurality of joint arms arranged in a straight line and hinged in sequence, wherein the two ends of the joint arm farthest from the proximal end are respectively provided with a walking wheel and a gripping foot, both can support and drive the wheel-foot conversion multi-legged robot, and the robot realizes wheel running and foot running mode conversion by switching the use of the walking wheel and the gripping foot of the multi-joint leg. The multi-legged robot designed by the utility model can be switched between the use of the walking wheel and the gripping foot, and then the wheel running and foot running mode conversion is realized, and the innovative design significantly enhances the obstacle crossing ability of the robot.

The embodiment of the application provides a cleaning robot and an obstacle crossing method, wherein the cleaning robot comprises a body, a gravity center adjusting mechanism, and a sewage recovery assembly for recovering a wet cleaning element of the body, the sewage recovery assembly comprises a sewage tank and an air pump, the gravity center adjusting mechanism comprises a counterweight cavity and a counterweight element movably arranged in the counterweight cavity; the air pump is communicated with the wet cleaning element through the sewage tank at an air inlet end, and is communicated with the counterweight cavity at an air outlet end. By reusing the air pump of the cleaning robot, the counterweight element is driven to move, so that the gravity center of the body is adjusted, and the obstacle crossing capability of the robot is improved; meanwhile, the need for additionally arranging an independent driving device is avoided, the structure of the robot is simplified, and the manufacturing cost and design complexity are reduced.

A traveling wheel mechanism and a cleaning device. The traveling wheel mechanism (100a, 100b) is mounted on a device body (300); the traveling wheel mechanism (100a, 100b) comprises a traveling wheel assembly (10) and a power assembly (20); the traveling wheel assembly (10) is rotatably connected to the device body (300); and the power assembly (20) can selectively extend and retract, can abut against the traveling wheel assembly (10) or the device body (300), and can slide relative to the traveling wheel assembly (10) or the device body (300). The obstacle crossing height, the escape capability, and the adaptability to different road surfaces of the device comprising the traveling wheel mechanism (100a, 100b) are improved.

The utility model discloses a chassis lifting structure of a cleaning robot and the cleaning robot, the chassis lifting structure comprises a main body, a driven wheel assembly and a lifting mechanism, the driven wheel assembly is arranged at the bottom of the main body, the driven wheel assembly is arranged close to the head of the main body, and the driven wheel assembly comprises a mounting frame and a driven wheel; a guide wheel is arranged on the front side of the mounting frame, and the driven wheel is rotationally connected to the mounting frame; the lifting mechanism is arranged on the main body, is in transmission connection with the mounting frame and is used for driving the mounting frame to lift so as to adjust the head height of the main body. When the cleaning robot encounters an obstacle, the driven wheel assembly can be driven by the lifting mechanism to adjust the head height of the main body, lifting of the shell can be driven by lifting of the main body, meanwhile, the obstacle crossing action of the cleaning robot is smoothly completed through guiding of the guide wheels, and the obstacle crossing success rate of the cleaning robot is increased.

The application discloses a kind of humanoid robot motor skills training methods based on body-object-environment relationship modeling.First, the reference motion data of target motor skills and interactive object, environment state data are obtained;During the training process, the body-object, body-environment, object-environment relationship characteristics and contact state are calculated based on the robot ontology, object and environment state, and the relationship enhanced state representation is constructed;Input it into the adversarial discriminant network to generate the adversarial reward representing the naturalness of action and the consistency of interaction;Finally, the adversarial reward, task, stability and safety reward are fused, and the control strategy is trained based on reinforcement learning.Compared with existing methods, the application explicitly learns the interaction relationship, reduces the complexity of reward engineering, improves the naturalness of movement, the rationality of interaction and the environmental adaptability, and is suitable for whole-body interaction tasks such as carrying, supporting and obstacle crossing.