50results about How to "High degree of anthropomorphism" patented technology

The invention provides an under-actuated self-adaptive anthropomorphic manipulator. The under-actuated self-adaptive anthropomorphic manipulator comprises a shell module, a four-finger module, a thumb module, a transmission module and a self-locking module, wherein the four-finger module, the thumb module, the transmission module and the self-locking module are arranged in the shell module. The four-finger module comprises an index finger, a middle finger, a ring finger and a little finger. The tail end of each finger is connected with the transmission module through a flat key. The thumb module is directly connected with the palm. The transmission module and the self-locking module are both arranged in the palm. The transmission module comprises a sliding block set, gear and rack mechanisms and a connecting rod mechanism, wherein the sliding block set slides backwards under the effect of the pulling force of a pulling rope, the gear and rack mechanisms on the two sides are driven to move in an engaged mode, racks drive the connecting rod mechanism to rotate, and then driving force is transmitted to the finger module to complete the grabbing motion. A pawl mechanism in the self-locking module is engaged with ratchets on a ratchet plate after grabbing, so that the grabbing state is maintained. According to the under-actuated self-adaptive anthropomorphic manipulator, self-adaptive wrapping and grabbing of objects with certain sizes can be achieved, and the application range is wide.

The invention discloses a disc-tooth switching multi-mode integration self-adaptive robot finger device, and belongs to the technical field of robot hands. The robot finger device comprises a base, two finger sections, two joint shafts, two motors, two driving pressure plates, two driven pressure plates, a sliding pressing block, multiple sets of transmission mechanisms, limiting blocks, spring pieces and the like. According to the robot finger device, multi-grabbing mode integrating coupling self-adaption, horizontal clamp self-adaption and a locking enhancing function finger are realized; according to the different shapes, sizes and positions of objects to be grabbed, and the device is provided with three basic grabbing modes of coupling grabbing, horizontal clamping grabbing and self-adaptive grabbing and three combined grabbing modes of coupling self-adaptive grabbing, horizontal clamp self-adaptive grabbing and locking enhancing grabbing; the grabbing modes can be freely and electrically switched; during switching, no idle running state exists so that the posture of fingers in the switching process can be kept; and the device is high in grabbing efficiency, wide in range, compact in structure, easy to control and suitable for various robots needing to grab objectsc.

The embodiment of the invention discloses an emotion information processing method and device, computer equipment and a readable storage medium, which is beneficial to displaying the emotion dissipation process of a robot and improving the humanoid degree of the robot. The method provided by the embodiment of the invention comprises the steps of determining first emotion information according to afirst emotion triggering event in response to the fact that a robot detects the first emotion triggering event; generating first control information according to the first emotion information, wherein the first control information is used for indicating the robot to sequentially execute a first behavior and a second behavior for expressing emotion, the first behavior is used for expressing an emotion indicated by the first emotion information, and the second behavior is used for expressing an emotion which is lighter than the emotion indicated by the first emotion information.

The invention discloses an outdoor targeting robot. The outdoor targeting robot comprises a chassis; a target body capable of rotating is mounted on the chassis; a skeleton, a base plate, a steering mechanism, a driving mechanism, a controller mounting plate, a wireless communication unit, a controller, a power supply body and an autonomous navigation device are mounted in the chassis; the drivingmechanism, the steering mechanism, the controller and the wireless communication unit are electrically connected with the power supply body; the controller is fixedly mounted on the controller mounting plate; the controller is in control connection with the autonomous navigation device, the steering mechanism and the driving mechanism; and the controller is in communication connection with an upper control center through the wireless communication unit. The outdoor targeting robot adopts an autonomous navigation technology to achieve such functions as path planning and autonomous obstacle avoidance, controls a robot driving mechanism and the steering mechanism to autonomously move from an initial point or a target point to a next target point or terminal according to specific scene conditions, realizes intelligent movement and control, and improves the training personification degree.

The invention discloses a translational motion idle-stroke transmission coupled and self-adaptive robot finger device and belongs to the technical field of robot fingers. The translational motion idle-stroke transmission coupled and self-adaptive robot finger device comprises a base, two finger sections, two joint shafts, a driver, a transmission mechanism, three gears, three racks, two spring pieces, two shifting levers and two convex blocks. The translational motion idle-stroke transmission coupled and self-adaptive robot finger device achieves the coupled and self-adaptive combined grabbing mode integrally, the two joints can be linked to pinch an object with the tail ends, the actions highly simulate human beings, and besides, the first finger section can be rotated to touch the object firstly, and then the second finger section is rotated to cover and hold the object, so that the effect of self-adaptive grabbing of objects with different shapes and dimensions is achieved. According to the device, transmission is precise and stable, and grabbing is stable and reliable; the two joints are driven through only one driver instead of complex sensing systems and real-time control systems; and meanwhile, the translational motion idle-stroke transmission coupled and self-adaptive robot finger device is simple in structure, short in transmission chain, small in number of parts, small in size, low in weight, easy to assemble and low in manufacturing cost.

The embodiment of the invention provides a virtual character generation method and device, electronic equipment and a storage medium. The method comprises the step of utilizing a motion state recognition model to identify the motion state of each joint based on the figure joint point information. Therefore, the virtual character is effectively simulated based on the motion state of the joint. Compared with the prior art, the scheme of the virtual character based on the acquired position information of the human body articulation points is adopted; according to the scheme, due to the fact thatthe motion state recognition model is used for recognizing the motion state, a good result can also be achieved for virtual character behavior simulation when part of joint information is lost, the virtual character is generated based on the motion state of each joint, and the personification degree of the virtual character is higher.

The invention discloses a multi-model intelligent robot system and a construction method, and the system comprises a digital operation service layer which comprises a strategy center, an operation center and basic services; the strategy center is used for scheduling and deciding to use various robot models to cope with different customer responses; the operation center is used for data analysis and model training of the multi-modal intelligent robot; the basic services are used for providing standard management function services of the digital operation service; the system also comprises an intelligent middle platform layer used for responding to the digital operation service layer and providing a plurality of robot models supporting semantic comprehension and task type multi-round voice dialogue services; a data service layer used for providing data storage and query services; and an interface service layer is used for integrating with a corresponding standard service interface for calling. The invention can improve the personification service capability of the intelligent robot, supports the multi-model service combination, achieves the efficient optimization operation of the robot, and is suitable for more business scenes.

The invention provides a game image processing method and device, equipment and a storage medium. The method comprises steps that feature information of a second virtual character which is located in the view field of a first virtual character in a game scene image and is closest to the center line position of the view field is obtained; a skill drop point position of the first virtual character in the game scene image is obtained; the neural network model is trained according to the game scene image, the feature information of the second virtual character and the skill drop point position, and the feature information of the second virtual character and the skill drop point position serve as label data of the game scene image; a technical problem of how to introduce artificial intelligence players into asymmetric confrontation competition games is solved.

The invention provides a virtual character driving method, device and equipment based on expression recognition, and relates to the fields of artificial intelligence, deep learning, machine learning, virtual reality and the like in the computer technology. According to the method, the face image of the user is obtained in real time in the conversation between the virtual character and the user, and the target classification of the current expression of the user is accurately recognized according to the face image of the user through the base model and the multi-modal alignment model for face expression recognition; and when it is determined that the target classification belongs to a preset expression classification and currently meets a response triggering condition of the target classification, determining corresponding driving data according to a response strategy corresponding to the expression classification of the user expression, and driving the virtual character to execute a corresponding response behavior according to the driving data and a three-dimensional image rendering model of the virtual character. Therefore, the virtual character can make a timely response to the expression of the user, the personification degree of the virtual character is improved, and the interaction between the virtual character and people is smoother and more intelligent.

The invention discloses an electronic standardized patient simulation system for physical examination training and examination. The system comprises a human body model and an electronic system, the human body model is provided with a plurality of operation points, observation points and an auscultation or palpation area, and the electronic system is electrically connected with the operation points, the observation points and the auscultation or palpation area. According to the invention, multiple test items are integrated on one human body model, so that the human body model is convenient to carry and store; the anthropomorphic degree is high, and the system is suitable for training and examination; and the model is electronized, so that the positive characteristics of the model can be converted, and the universality is higher.

The invention provides a method and system for constructing a three-dimensional space based on voxel data, a server and a computer readable storage medium. According to the invention, pixel data is converted into voxel data so as to be highly matched with GPU operation; layer data and connection data are designed according to the voxel data, and a parallel computing speed is higher in virtue of the layer data and the connection data, so compared with an existing CPU operation mode, the performance of a data operation scheme based on GPU is 2-3 orders of magnitude higher, and an AI algorithm is finer and higher in anthropomorphic degree; in a dynamic updating process, due to the fact that the parallel computing speed is high, the data of a global super-large map of a game can be rapidly updated; and the whole set of data is divided into static data with high memory occupation and dynamic data with low memory occupation, so video memory space occupied in the whole game process is small, time for refreshing the dynamic data each time is extremely short, and the game is smoother.

The invention relates to an intelligent automobile following decision and control method based on cognitive risk balance. The method comprises the steps: obtaining a cognitive risk based on the time headway and reverse collision time when an intelligent automobile is in a following working condition; judging whether the risk is within a cognitive risk balance interval or not, if not, predicting the response acceleration of the human driver under the action of the corresponding risk intensity through a function between the longitudinal acceleration and the cognitive risk, taking the response acceleration as anthropomorphic response acceleration of the intelligent automobileto deal with the corresponding risk, and using the anthropomorphic response acceleration to adjust the cognitive risk; keeping the balance; and based on the response acceleration, acquiring the throttle opening and the brake pressure intensity through prediction of an automobile longitudinal inverse dynamic model so that the automobile speed is adjusted in real time, and intelligent automobile following is achieved. Therefore, the application of an anthropomorphic cognitive risk balance mechanism to the longitudinal driving behavior decision of the intelligent automobile is realized, and the driving decision rule of an excellent driver is learned to guide the anthropomorphic automobile following driving of the intelligent automobile.

The invention relates to the technical field of artificial intelligence, and discloses a speech synthesis model training method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining original text data and a speaker identifier, processing the original text data, obtaining an original text vector and a corresponding original phoneme vector, and carrying out feature enhancement processing on the original text vector and the original phoneme vector to obtain a target text vector and a target tone vector which are more significant in vectors; based on the speaker identifier, obtaining a corresponding target voiceprint vector, splicing the target voiceprint vector, the target text vector and the target tone vector, and training a speech synthesis model by using the spliced target hidden vector to obtain a target speech synthesis model corresponding to the speaker identifier, therefore, the similarity between the voice data synthesized by the updated target voice synthesis model and the voiceprint of the speaker is improved, and the personification degree of the target voice synthesis model is increased.