105 results about "Robot perception" patented technology

A robot includes a hazard sensor, a locomotor, and a system controller. The robot senses a hazard intensity at a location of the robot, moves to a new location in response to the hazard intensity, and autonomously repeats the sensing and moving to determine multiple hazard levels at multiple locations. The robot may also include a communicator to communicate the multiple hazard levels to a remote controller. The remote controller includes a communicator for sending user commands to the robot and receiving the hazard levels from the robot. A graphical user interface displays an environment map of the environment proximate the robot and a scale for indicating a hazard intensity. A hazard indicator corresponds to a robot position in the environment map and graphically indicates the hazard intensity at the robot position relative to the scale.

Within a moving robot of narrow footprint, having quick traveling performance on a plane-surface, as well as, anti-tumbling function, and further being suitable for operations under coexistence with a human being, being able to travel coping with traveling situations on a level difference, etc.: in front and rear of main driving wheels 2 and 3, each being controlled through the inverted pendulum control, are disposed supporting legs 4 and 5, tips of which can be lifted up and down, wherein the tips of the supporting legs 4 and 5 are positioned to keep a predetermined distance between a traveling surface, when running on the inverted two-wheels travel, and the supporting legs 4 and 5 are fixed or either one in the fall-down direction is thrust out into the fall-down direction, so as to protect it from falling down. Further, upon the basis of detection information of floor-surface distance sensors 4e and 5e and side-surface distance sensor 4f, 4g, 5f and 5g, which are provided at the tips of the supporting legs 4 and 5, the robot senses an existence of a level difference and / or an inclined surface, so as to let the supporting legs to escape from the level difference and / or the inclined surface, and holds the position of gravity center thereof, stably, through other one of the supporting legs, being landed on the ground, and the main driving wheels 2 and 3; thereby enabling to travel over the level difference and the inclines surface.

The invention discloses a mine intelligent robot inspection system and method. The mine intelligent robot inspection system comprises a track circulating rotation subsystem, a robot sensing subsystem,a wireless charging subsystem, and a remote control and deep learning subsystem. The track circulating rotation subsystem is used for driving the robot sensing subsystem to rotate circulatively around the track. The robot sensing subsystem is used for detection on the surrounding environment and communication. The remote control and deep learning subsystem is for receiving the detected data, andsending a control command to the track circulating rotation subsystem and the robot sensing subsystem. Through the technical scheme of the invention, under the premise of complying with the explosion-proof standard of a coal mine, according to the collection of environmental parameters, the deep learning is used to predict the unknown risk, and the remote control is used to control the robot inspection system, so that the automatic wireless charging is realized, the production safety is improved and the reliability of the inspection process is ensured.

Within a moving robot of narrow footprint, having quick traveling performance on a plane-surface, as well as, anti-tumbling function, and further being suitable for operations under coexistence with a human being, being able to travel coping with traveling situations on a level difference, etc.: in front and rear of main driving wheels 2 and 3, each being controlled through the inverted pendulum control, are disposed supporting legs 4 and 5, tips of which can be lifted up and down, wherein the tips of the supporting legs 4 and 5 are positioned to keep a predetermined distance between a traveling surface, when running on the inverted two-wheels travel, and the supporting legs 4 and 5 are fixed or either one in the fall-down direction is thrust out into the fall-down direction, so as to protect it from falling down. Further, upon the basis of detection information of floor-surface distance sensors 4e and 5e and side-surface distance sensor 4f, 4g, 5f and 5g, which are provided at the tips of the supporting legs 4 and 5, the robot senses an existence of a level difference and / or an inclined surface, so as to let the supporting legs to escape from the level difference and / or the inclined surface, and holds the position of gravity center thereof, stably, through other one of the supporting legs, being landed on the ground, and the main driving wheels 2 and 3; thereby enabling to travel over the level difference and the inclines surface.

The invention discloses a method for realizing autonomous exploration mapping and autonomous path covering of an indoor robot cleaner. The method comprises the following steps: 1) the robot cleaner senses the current area environment, obtains the depth information of the current shooting area, and determines a boundary position; 2) the robot moves autonomously to the position, completes the scanning and map generation of the upper area, and senses the environment of the new area; 3) the robot cleaner finally completes the generation of an indoor environment complete map through the continuous boundary exploration and new regional scanning, and distinguishes the area that cannot be passed and the area that can be passed in the map; 4) the indoor environment map can be completed, and the map area can be decomposed to generate the cleaning path for covering each area; and 5) a motion control module drives the robot cleaner to finally complete the cleaning of the indoor environment in accordance with the planned area order and path. The method overcomes the shortcomings of low coverage and high cleaning repetition rate of a conventional robot cleaner, thereby improving the robustness of the system and the work efficiency.

The invention discloses a robot monitoring and automatic mobile system operation method based on distributed intelligent monitoring controlling nodes. A robot monitoring and automatic mobile system based on the distributed intelligent monitoring controlling nodes is built and comprises the intelligent monitoring controlling nodes and a server; a real environment map is drawn and matched with real map coordinates; the server filers and blends image collection result data of all the intelligent monitoring controlling nodes in each cycle; the filtered and blended result data is used for generating a real-time control instruction of the robot and distributing the instruction to a corresponding intelligent monitoring controlling node. The method has the advantages of greatly reducing robot body manufacture cost and achieving robot dead-corner-free monitoring due to the fact that a widely used monitoring camera replaces traditional robot induction devices such as a laser and a speedometer, adopting the distributed structure, greatly improving system expandability and achieving expansion of a monitoring range due to the fact that the number of the nodes can be conveniently increased and reduced at will.

The invention discloses a robot simulated navigation method based on rat brain-hippocampal navigation, and belongs to the technical field of robot motion navigation. The method is characterized by that a robot CPU (Central Processing Unit) predefines a grid cell layer cell layer G and a place cell layer P by imitating rat brain and hippocampal tissues, motion information including a motion direction phi and a velocity v is collected when a robot explores an environment, the collected motion information is input into the grid cell layer, the motion information is integrated by using a shock interference model in the grid cell layer to form a grid cell layer response, a connection weight value W of the grid cell layer cell layer G and a place cell layer P is calculated by using a neural network algorithm to form a location cell-grid cell response network, and the robot perception for a space namely is stored in the form of a neural network response. The robot controls a motion decision module to implement motion decision-making according to space information stored in a cognitive map, and the navigation method disclosed by the invention independently obtains a better spatial cognitive effect during an exploration motion of the robot, and can be applied to industrial robots, service robots and the like.

A system, apparatus and application for providing robots with the ability to intelligently respond to perceived situations are described. A knowledge database is assembled automatically, based on distributed knowledge capture. The knowledge base embodies the “common sense,” that is, the consensus, of the subjects who contribute the knowledge. Systems are provided to automatically preprocess, or “clean” the information to make it more useful. The knowledge thus refined is utilized to construct a multidimensional semantic network, or MSN. The MSN provides a compact and efficient semantic representation suitable for extraction of knowledge for inference purposes and serves as the basis for task and response selection. When the robot perceives a situation that warrants a response, an appropriate subset of the MSN is extracted into a Bayes network. The resultant network is refined, and used to derive a set of response probabilities, which the robot uses to formulate a response.

The invention provides a robot perception and understanding method based on man-machine collaboration. The method allows an operator to help robot to perceive and understand the environment. The method comprises the steps of firstly, audio visual mode perception natural interaction, secondly, target describing and understanding and thirdly task teaching and learning. Intelligence of the human is used, the most natural and efficient interaction manner of the human is used for man-machine communication and assistance, and the robot perception and understanding flexibility, intelligence and adaptation are improved.

The invention discloses an incomplete point cloud completion method based on hidden space topological structure constraints, and belongs to the technical field of computer vision. On the basis of a three-dimensional shape complementing technology of the single-view incomplete point cloud, complementing the single-view incomplete point cloud obtained by converting the depth map to obtain a complete three-dimensional point cloud shape of the object, and realizing rapid reconstruction of geometric information of the perceived object. S1, collecting three-dimensional space coordinates of a target object; S2, performing point cloud attitude normalization on the original incomplete point cloud input by the system; S3, predicting a corresponding complete key point skeleton based on the attitude normalized incomplete point cloud; S4, based on the predicted complete key point skeleton, recovering a dense complete point cloud corresponding to the predicted complete key point skeleton; and S5, carrying out optimization of robot sensing task based on the complete three-dimensional point cloud. According to the method, shape completion can be carried out on the real three-dimensional point cloud of any pose, the generalization and robustness are higher, the application range is wide, the speed is high, and the anti-noise capability is high.

The invention provides an obstacle-crossing robot for transmission line operation and an obstacle-crossing method thereof. The robot comprises a body, a front swinging arm, a rear swinging arm, a front roller wheel, a rear roller wheel, a front arm slider, a rear arm slider and an auxiliary arm. The method comprises the following steps: when a sense director of the robot detects that obstacles exist, the robot first loosens conductors in wheel-type hands of the two front arms, and then swings the two arms outward, moves round the obstacles by 180 degrees of rotation and grips the conductors again, at the moment, the robot crosses the obstacles. The two rear arms adopt the same method to cross the obstacles. After crossing the obstacles, the robot reverses the four arms to prepare for the next obstacle-crossing. The robot is suitable for the signal conductor and bundled conductor, and can complete multiple operations. The method features clear and simple steps. All the driving motors are all in sequential working state, without multi-motor linkage. The robot is characterized by considerably easy control system manufacturing and programming, thus having good adaptability to the field.

The present invention discloses one kind of robot behavior transmission method based on network function and relates to network robot technology. The technological scheme includes for the sending end robot to sense the user's behavior via the sensor, convert it into behavior signal, perform signal data analysis to obtain the behavior mark, and send the behavior mark through the network to the receiving end robot; and for the receiving end robot to receive the behavior mark, analyze and exhibit behavior. The present invention also discloses the unit for the robot behavior transmission. The present invention has increased information transmission mode, effective emotion transmission and comprehensive interaction.

The invention relates to a three-dimensional tactile force detection method based on micro-vision, and belongs to the technical field of intellisense of robot dexterous hand executing tail ends. The three-dimensional force detection technology method for verification is designed on the basis of an independently-designed tactile sensor, a good effect is achieved in the gripping process, and the method belongs to the technical field of robot perception. When the sensor directly makes contact with an object, an elastomer can be subjected to force to generate deformation, marking points can correspondingly generate displacement, and image information generated before and after the marking points generate displacement is captured by an image collecting camera. Deviation information of the marking points is calculated through comparison of the image information generated before and after the force is applied, and then a regressed three-dimensional force prediction vector is obtained through neural network training and predicting. According to the three-dimensional tactile force detection method based on micro-vision, three-dimensional force tactile sense information, three-dimensional force slip sense information and the like in the dexterous hand gripping process can be detected, and the method can be applied to the field of tactile perception of robot dexterous hands.

The invention relates to a method for identifying the grabbing stability of a manipulator based on deep learning, and belongs to the technical field of robot perception. The method in the invention comprises the steps of: acquiring tactile data in different grabbing stabilities at first, then, converting time series picture data into a picture, training a deep learning network by utilizing tactile picture data, and finally, identifying the tactile picture data having the unknown stability by utilizing the deep learning network. By means of the method in the invention, training of the deep learning network is carried out based on original data different in time duration, so that the obtained network has the robustness in data duration. The deep learning network in the method is used for updating network parameters based on integrated input data; and influence on the network is very low only when labels of a few data have errors. Thereby, the method has the robustness to noise data.

The invention discloses a whisker sensor suitable for an inclined angle between robot perception and a barrier, comprising a laser, a one-dimensional PSD sensor, a whisker, a base and a shell. The whisker sensor belongs to the detection device part of a robot and is arranged on a robot body, and an optical position provided with the whisker sensor is at the front end of the robot body; when the robot body is close to the barrier, the whisker sensor realizes the collection of the inclined angle theta between the whisker and the barrier by the one-dimensional PSD sensor and a laser source arranged on the root part of the whisker and then obtain the inclined angle theta between the machine body and the barrier according to the advancing amount S and the tail end displacement amount W of the machine body; in addition, the whisker sensor can realize the barrier avoidance by turning away from the inclined angle theta.

The invention discloses a bionic-based robot perception control system and a control method. The system comprises a robot body, a mechanical arm and a visual inertial navigation module, the robot bodyis provided with a mechanical arm base, the mechanical arm is mounted on the mechanical arm base, the mechanical arm is provided with a plurality of rotating joints, one steering gear is arranged ineach rotating joint, the visual inertial navigation module is installed at the end of the mechanical arm, and the visual inertial navigation module uses a visual inertial odometer to establish a cloudmap synchronously in real time. The system adopts the visual inertial navigation module as a visual navigation and positioning system of a bionic robot, utilizes a multi-degree-of-freedom movement ofthe mechanical arm to make a robot visual system free from the limitation of the movement of the robot body so that the robot can cope with complex and varied dynamic scenes, utilizes a visual inertia SLAM technology to achieve real-time navigation and positioning functions, which makes the bionic robot have a strong human-computer interaction function, and can be applied in the various dynamic scenes.

The invention discloses a multi-robot pollution control method based on Voronoi diagrams. The multi-robot pollution control method comprises the following steps: robot sensing, task allocating, target tracking, region covering and obstacle avoiding. A system is divided into two layers: a virtual layer and a real layer; wherein in the virtual layer, a Voronoi sensing region is divided by a weighed distance so as to improve the monitoring, identification and communication efficiency of the system, thereby facilitating the reconstruction of an ambient network; the advantages and disadvantages of each task over system pollution treatment are weighed by distributed robots through the utilization of weight values in combination with self conditions and pollution source leakage conditions, and the tasks are decided by the robots selves; in the real layer, when the robots meet an obstacle, a virtual target point is updated into the edge point of the obstacle in combination with the virtual layer; neutralized substances are released according to pollution degree in the Voronoi region. The control method has the advantages that formation patterns can be arbitrarily changed according to dynamic environments so as to ensure that the robots can quickly, stably and efficiently treat pollutants and avoid the obstacle, and the independence and the self-adaptability of the multi-robot system are improved.

The invention provides a robot interactive behavior control method, a robot interactive behavior control device and a robot. The control method includes the following steps that: information sensed by the robot is acquired; at least one pre-defined sensing unit generates sensing data containing the identifiers and values of the sensing units according the sensed information; a control entry which is matched with the generated sensing data and is used for responding to the information sensed by the robot to control the behaviors of the robot is searched, wherein the control entry contains a triggering condition formed by the at least one sensing unit and behaviors triggered by the triggering condition; and if the control entry matched the generated sensing data is found, the robot is made to perform the behaviors in the found control entry. According to the robot interactive behavior control method, the robot interactive behavior control device and the robot of the invention, the sensing unit is defined as a minimum unit for controlling the interactive behaviors of the robot; and the control entry is set according to the sensing unit so as to control the interactive behaviors of the robot, and therefore, the adaptive interactive behavior ability and intelligent degree of the robot can be effectively improved.

The invention discloses a robot orientation and hand-eye relationship simultaneous calibration method based on bundle adjustment, which comprises the steps of performing three-dimensional reconstruction on a detection target through a monocular stereoscopic vision algorithm to obtain a camera external parameter matrix without a scale factor alpha; utilizing a robot kinematics equation to obtain arobot motion orientation matrix; representing a calibration equation A(alpha)X=ZB by using a matrix direct product parameter method, and solving a robot orientation and hand-eye relationship initial value; and carrying out iterative optimization on the robot orientation and hand-eye relationship initial value and a three-dimensional reconstruction result obtained by the monocular stereoscopic vision algorithm by using a beam adjustment optimization model. According to the robot orientation and hand-eye relationship simultaneous calibration method based on beam adjustment, based on the light beam adjustment model, accurate three-dimensional positioning can be performed on the target while the robot orientation and the hand-eye relationship are solved, accurate positioning of the detection target is also realized while dependence on the target is eliminated, and thus the intelligent robot is endowed with sensing and positioning functions.

There is disclosed a computer-implemented method of handling an audio dialog between a robot and a human user. The method includes: during the audio dialog, receiving audio data and converting the audio data into text data; in response to the text data, determining a dialog topic, the dialog topic comprising a dialog content and a dialog voice skin, and the dialog content comprising a plurality of sentences; determining a sentence to be rendered in audio by the robot; and receiving a modification request of the determined dialog sentence. Described developments for example comprise different regulation schemes (e.g. open-loop or closed-loop), the use of moderation rules (centralized or distributed) and the use of priority levels and / or parameters depending on the environment perceived by the robot.

The invention relates to smart home and discloses a smart home system with robot thought characteristics in a home environment. The smart home system comprises two parts, namely a robot sensing and acting module (I) and a robot intelligence thought module (II), wherein the robot sensing and acting module consists of an air quality detection circuit (1), a humidity detection circuit (2), a temperature detection circuit (3), an illumination detection circuit (4), a vibration detection circuit (5), an infrared detection circuit (6), a smoke detection circuit (7), a sound detection circuit (8), a central processing unit (9), a communication circuit (10), a voltage stabilizing circuit (11), an audio input circuit (12), a sound control circuit (13), a sound play circuit (14), a full-color light-emitting diode (LED) illuminating circuit (15) and an anion generating circuit (16); and the robot intelligence thought module (II) consists of a bus interface circuit (17), an ARM hardware circuit (18), an Android operating system (19), a smart home application program (20) and a TV output circuit (21). According to the comprehensive sensing state and manual intelligent thought control, the home is safe, convenient, comfortable and warm.

The invention discloses a leg-foot type intelligent star surface detection robot sensing system and a working method thereof. The leg-foot type intelligent star surface detection robot sensing systemcomprises a three-dimensional laser sensor, a left-eye visible light camera, a right-eye visible light camera, a structured light measurement camera, an inertia measurement unit and an industrial personal computer. The working method comprises the following steps of: 1) integrating the three-dimensional laser sensor, the structured light measurement camera and the inertia measurement unit to realize the functions of three-dimensional reconstruction of complex topography and landform of a star surface, multi-level semantic topological map construction, robot positioning, real-time detection ofstatic / dynamic obstacles, path optimization based on robot motion constraint, trajectory tracking and the like; and 2) and centralizing the binocular vision cameras for carrying out high-precision three-dimensional pose measurement on a to-be-collected sample. The leg-foot type intelligent star surface detection robot sensing system not only can assist the foot-type robot to autonomously adapt todifferent terrain environments such as soft and hard terrain environments, realize long-distance, intelligent obstacle avoidance and autonomous roaming and safely arrive at a pre-specified target detection position, but further can assist a mechanical arm tail end tool to perform refined operation on a collected sample.

The invention provides a piezocapacitance and piezoresistance coupled proximity sensing and contact force sensor, belongs to the technical field of tactile safety, and particularly relates to a piezocapacitance and piezoresistance coupled proximity sensing and contact force sensor. The problems that a tactile sensor with a single detection principle is difficult to realize measurement of large-range force change, the response speed is difficult to break through the millisecond magnitude, and the contact sensing function before collision between an object and the sensor is lost are solved. Thepiezoresistive sensor comprises a flexible substrate, coplanar electrodes, piezoresistive elastomers and surface bulges, wherein the flexible substrate is located at the bottom, the coplanar electrodeis arranged on the flexible substrate, the piezoresistive elastomer is of a cavity structure, the piezoresistive elastomer is arranged above the flexible substrate and the coplanar electrode, an airhole is formed in an area defined by the piezoresistive elastomer and the flexible substrate, and the surface protrusion is arranged on the upper surface of the piezoresistive elastomer. The robot ismainly used for sensing external environment changes.

The invention belongs to the technical field relevant to flexible robot sensing control, and discloses a force position feedback control system for a flexible robot. The system comprises a main control circuit board and a strain sensor which are respectively arranged on the flexible robot; a Wheatstone bridge, a force position sensing module and a driving control module are arranged on the main control circuit board; the force sensor module is connected with the driving control module; the driving control module is connected with a motor of the flexible robot; the Wheatstone bridge is electrically connected with the strain sensor and the force sensing module; the force sensing module comprises a shape sensing module for obtaining the deformation position according to the offline storage strain mode and the received strain signal, and a contact force sensing module for solving the deformation position contact force through the deformation position shape and the motor pulling force. Theshape and stress sensing of the flexible robot are realized through the strain; through the closed loop control, the accurate shape and position control are realized through closed loop control; the flexibility is high; the integration degree is high.

The invention belongs to the field of robot perception, and particularly relates to a disparity map enhancement method based on RGB and DVS image fusion in a high dynamic range scene. The method comprises the following steps: S1, deploying a binocular RGB camera and a DVS camera, and calibrating the binocular RGB camera and the DVS camera; s2, collecting an RGB image and a DVS image of a binocularcamera in a scene, and performing multi-scale weighted fusion after registration; s3, generating an HDR image for computer vision for the fused image; and S4, based on the HDR image generated in thestep S3, using an improved binocular stereo matching algorithm SGM to generate a disparity map. In a scene with a large imaging dynamic range such as a tunnel, the problems of under exposure and highexposure of a camera are solved, the quality of a generated image is improved, and meanwhile, for the problems of discontinuity and instability of an image edge region, edge detail information is enriched as much as possible by introducing other information sources, and the accuracy of the finally generated disparity map at the edge of the image is improved.

A foot-type robot terrain perception method based on a virtual sensor belongs to the field of robot perception, and comprises the following steps: S1, establishing a ground contact detection neural network model and a soil classification machine learning model; S2, under the conditions of different terrains and different gaits, collecting the angle of a leg joint, the angular speed of the leg joint, the motor current and the contact force data of the leg and the ground of the foot-type robot as samples; S3, training a ground contact detection neural network model and a soil classification machine learning model by using the samples collected in the S2; and S4, taking the ground contact detection neural network model and the soil classification machine learning model trained in the S3 as aterrain perception system of the foot-type robot for terrain perception. According to the method, the walking stability and the movement capacity of the robot can be improved, and meanwhile the robustness of the robot and the reliability of behaviors of the robot are enhanced; in addition, hardware of the robot is simplified, and the design, machining and maintenance cost is reduced.

A system and method comprises marking or identifying an object to be perceptible to robot, while being invisible or substantially invisible to humans. Such marking can facilitate interaction and navigation by the robot, and can create a machine or robot navigable environment for the robot. The machine readable indicia can comprise symbols that can be perceived and interpreted by the robot. The robot can utilize a camera with an image sensor to see the indicia. In addition, the indicia can be invisible or substantially invisible to the unaided human eye so that such indicia does not create an unpleasant environment for humans, and remains aesthetically pleasing to humans. For example, the indicia can reflect UV light, while the image sensor of the robot can be capable of detecting such UV light. Thus, the indicia can be perceived by the robot, while not interfering with the aesthetics of the environment.

The present invention provides a robust slam method for indoor robots combined with environmental semantics. The method comprises: indoor robot positioning of the fusion of GPS global signals, IMU inertial measurement unit signals and lidar signals; an indoor robot perception of the fusion of radar data image data, radar sensor calibrated external parameters, pre-camera calibrated external and internal parameters, and a host speed and angular velocity; an indoor robot semantic map of the fusion of the NavigationInfo of a dreamview module, a LaneMarker from a sensing module, and positioning information of a positioning module; decision making of the indoor robot of the fusion of obstacle information, vehicle status, traffic lights, and map information; and indoor robot path planning for thefusion of positioning, a perception, prediction, routing, and a map. According to the method provided by the present invention, the robot can be more robust to the environment in motion and decisionmaking in the environment.

A system, apparatus and application for providing robots with the ability to intelligently respond to perceived situations are described. A knowledge database is assembled automatically, based on distributed knowledge capture. The knowledge base embodies the “common sense,” that is, the consensus, of the subjects who contribute the knowledge. Systems are provided to automatically preprocess, or “clean” the information to make it more useful. The knowledge thus refined is utilized to construct a multidimensional semantic network, or MSN. The MSN provides a compact and efficient semantic representation suitable for extraction of knowledge for inference purposes and serves as the basis for task and response selection. When the robot perceives a situation that warrants a response, an appropriate subset of the MSN is extracted into a Bayes network. The resultant network is refined, and used to derive a set of response probabilities, which the robot uses to formulate a response.

The invention provides a robot sensing device and a system. The robot sensing device comprises a cognitive module and a control module. The cognitive module is used for acquiring target information and transmitting the cognized target information to the control module in real time; and the control module is used for generating a control signal according to the information collected by the cognitive module, and the control signal is used for controlling a mechanical arm to grab a workpiece to be grabbed and monitor the working state of the mechanical arm in real time. The robot sensing device and the system provided by the invention can be mounted at the front end of the robot or fixed to a non-occluded target workpiece, and acquires an image, and the outline, the size, the position, the distance, the color and the perceived force feedback of a target object observed by the robot in real time, so that the robot obtains cognitive ability, can grab or assemble and detect in different positions, different dimensions, and the proper strength according to different workpieces.