447 results about "Motion space" patented technology

The invention relates to an external skeleton robot, in particular to a human-simulated external skeleton robot assisting the lower limbs. The human-simulated external skeleton robot assisting the lower limbs aims to solve the problems that an existing external skeleton robot is low in coupling degree of motion space and poor in wearing comfort, reliability and adaptation, and power needed by a motor is large. The human-simulated external skeleton robot assisting the lower limbs comprises an upper body back part, a left leg and a right leg. The left leg and the right leg respectively comprise a hip drive system, a knee drive system and a foot wearing system. A rear side connection board of the waist is in rotating connection with a load installation board. Each hip joint supporting board is provided with a first motor and a first reducer, wherein the first motor is provided with an encoder, and the output end of the first motor provided with the encoder is connected with the input end of the first reducer. Each hip joint connecting board can rotate in the vertical plane. Each thigh stretching board is in detachable connection with the corresponding hip joint connecting board. The output end of a main drive mechanism is connected with each crus connecting board. The lower surfaces of elastic boards are bonded with the upper surfaces of the rubber soles of the feet. The human-simulated external skeleton robot assisting the lower limbs can assist in walking.

A patient support apparatus for supporting a patient in a prone position during a surgical procedure is provided, including an open fixed frame suspended above a floor and a pair of spaced opposed radially sliding joints cooperating with the frame, each joint including a virtual pivot point and an arc of motion spaced from the virtual pivot point, the joints being movable along the arc providing a pivot ship mechanism for a pair of pelvic pads attached to the joints. A base for supporting and suspending a patient support structure above the floor, for supporting a patient during a surgical procedure, the base including a pair of spaced opposed vertical translation subassemblies reversibly attachable to a patient support structure, a cross-bar, and a rotation subassembly having two degrees of rotational freedom; wherein a location of each vertical translation subassembly is substantially constant during operation of the patient support structure.

The invention relates to the technical field of traffic control and artificial intelligence and provides a traffic signal self-adaptive control method based on deep reinforcement learning. The method includes the following steps that 1, a traffic signal control agent, a state space S, a motion space A and a return function r are defined; 2, a deep neutral network is pre-trained; 3, the neutral network is trained through a deep reinforcement learning method; 4, traffic signal control is carried out according to the trained deep neutral network. By preprocessing traffic data acquired by magnetic induction, video, an RFID, vehicle internet and the like, low-layer expression of the traffic state containing vehicle position information is obtained; then the traffic state is perceived through a multilayer perceptron of deep learning, and high-layer abstract features of the current traffic state are obtained; on the basis, a proper timing plan is selected according to the high-layer abstract features of the current traffic state through the decision making capacity of reinforcement learning, self-adaptive control of traffic signals is achieved, the vehicle travel time is shortened accordingly, and safe, smooth, orderly and efficient operation of traffic is guaranteed.

The invention provides an electric pressure cooker and an uncovering device thereof. The uncovering device includes a driving member and an uncovering button; the driving member is rotatably arrangedon an inner cover; the driving member is connected to a cooker cover and can drive the cooker cover to rotate; the uncovering button is rotatably arranged on the inner cover through a rotation shaft,and is connected to a housing; when the driving member rotates to a first preset position, a tooth of the cooker cover is separated from a tooth of the housing; and when the driving member rotates toa second preset position, the driving member is pressed against the uncovering button so as to drive the uncovering button to rotate with respect to the inner cover, and the uncovering button is separated from the housing. According to the uncovering device, when the driving member rotates to the second preset position, the driving member generates a transverse force acting on a second end of theuncovering button so as to drive the uncovering button to rotate, and then the transverse force generated by rotation of the driving member is converted into a transverse stroke, a desired motion space is small during a rotation process of the uncovering button, and the uncovering button can move smoothly.

The invention discloses a planning method for a mixed path of a mobile robot under a multi-resolution barrier environment. The invention aims to solve the problems of blindness of initial planning, environment modeling lack of flexibility and poor real-time obstacle avoidance capability of the present method. According to the technical scheme, a self-adapting inhomogeneous polar-radius polar coordinate modeling method is adopted for performing environment modeling on the motion space of the mobile robot; a gravity particle swarm searching method is adopted for planning an initial overall path from starting point to ending point; according to the initial overall path, a modified artificial potential field method is adopted for performing local dynamic obstacle avoidance by estimating the minimum safe distance and safe collision-preventing angle and for arriving at each initial overall path point in turn; and a final overall collision-free path is output after arriving the planning end point. According to the planning method provided by the invention, the blindness of initial overall planning and the environment modeling flexibility can be effectively improved, the real-time obstacle avoidance capability for dynamic unknown barrier is strong, and the method is high in speed, high in precision and strong in adaptability.

The invention provides a reinforcement learning based air combat maneuver decision making method of a UAV. A motion model of an airplane platform is created; principle factors that influence the air combat situation are analyzed; on the basis of the motion model and analysis on the air combat situation factors, a dynamic fuzzy Q learning model of air combat maneuver decision making is designed, and essential factors and an algorithm flow of reinforcement learning are determined; a state space of air combat maneuver decision making is fuzzified and serves as state input of reinforcement learning; typical air combat motions are selected as basic motions of reinforcement learning, and the triggering intensities of fuzzy rules are summed in a weighted manner, and a continuous motion space is covered; and on the basis of an established air combat dominant function, a return value of reinforcement learning is set in a rewards and punishment values weighing-superposing method. Thus, the autonomous maneuver decision making capability of the UAV during air combat can be improved effectively, the robustness is higher, an autonomous searching optimization performance is higher, and decisionsmade by the UAV are improved continuously in continuous simulation and learning.

A motion space presentation device includes: a work area generation unit configured to generate a three-dimensional region in which the movable robot operates; an image capture unit configured to capture a real image; a position and posture detection unit configured to detect an image capture position and an image capture direction of the image capture unit; and an overlay unit configured to selectively superimpose dither an image of a segment approximation model of the movable robot as viewed in the image capture direction from the image capture position, or an image of the three-dimensional region as viewed in the image capture direction from the image capture position, on the real image captured by the image capture unit, according to the difficulty in recognizing each image.

The invention discloses a collision avoidance planning method for mobile robots based on deep reinforcement learning in a dynamic environment, and belongs to the technical field of mobile robot navigation. The method of the invention includes the following steps of: collecting raw data through a laser rangefinder, processing the raw data as input of a neural network, and building an LSTM neural network; through an A3C algorithm, outputting corresponding parameters by the neural network, and processing the corresponding parameters to obtain the action of each step of the robot. The scheme of the invention does not need to model the environment, is more suitable for an unknown obstacle environment, adopts an actor-critic framework and a temporal difference algorithm, is more suitable for a continuous motion space while realizing low variance, and realizes the effect of learning while training. The scheme of the invention designs the continuous motion space with a heading angle limitationand uses 4 threads for parallel learning and training, so that compared with general deep reinforcement learning methods, the learning and training time is greatly improved, the sample correlation isreduced, the high utilization of exploration spaces and the diversity of exploration strategies are guaranteed, and thus the algorithm convergence, stability and the success rate of obstacle avoidance can be improved.

The invention discloses a six-leg walking robot capable of crossing obstacles. The walking robot comprises a left front leg (1), a right front leg (2), a left middle leg (3), a right middle leg (4), a left rear leg (5), a right rear leg (6) and a body (7). The body comprises a front body (7A) and a rear body (7B). The six legs are symmetrically disposed relative to the body (7). Each leg is provided with three joints, namely a hip joint, a thigh joint and a shank joint, coordination of the three joints guarantees front-rear swing, lifting, and bending and stretching of each leg. The three joints of each leg and body joints guarantee good obstacle crossing and slope climbing ability of the robot. The feet of each leg is connected in a buffer manner, and damping and energy accumulation recycling can be completed in a one-step cycle. The six-leg walking robot is flexible in motion, fast in response, large in motion space, small in motion inertia, and suitable for movements under complex non-structure topographic conditions of fields.

The invention relates to a vehicle-mounted movable type walk-assisting exoskeleton rehabilitation robot belonging to the technical field of rehabilitation engineering. The vehicle-mounted movable type walk-assisting exoskeleton rehabilitation robot comprises an auxiliary movable platform and a lower limb exoskeleton mechanism, wherein the lower limb exoskeleton mechanism is connected with the auxiliary movable platform; the auxiliary movable platform comprises an auxiliary movable trolley, a control cabinet and a suspension mechanism, and the auxiliary movable trolley is connected with the control cabinet and the suspension mechanism; the lower limb exoskeleton mechanism comprises a width-adjusting mechanism and two leg exoskeletons, the two leg exoskeletons are respectively connected to both sides of the width-adjusting mechanism, and the width-adjusting mechanism is connected with the auxiliary movable platform. The invention realizes more rehabilitation training functions by enlarging the motion space of the lower limb exoskeleton mechanism, also realizes the limitation protection effect on lower limb joints of human bodies, has compact structure and stable work, and can realize the functions of assisting walk, autonomous movement, lower limb rehabilitation training, and the like.

The invention relates to a discovery method of a track data hot spot based on local multilayer grids. The technical scheme includes dividing track motion space locally and in a multilayer mode according to distribution characteristics of track data and calculating cell densities; screening and expanding a density cell under a given threshold and calculating a candidate hot spot; and screening the hot spot based on the residence time of candidate hot spot track support counts and track data in the candidate hot spot. According to the method, self-adaptive effects of a divided grid coverage motion plane are good, suitable grid coverage motion can always be obtained in an iterative classification process as long as a sample point number threshold value in a cell is assigned, sample data are divided meticulously, initial classification parameters cannot affect discovery results greatly as long as a set density threshold is fixed, the method can be applied to mass data discovery, and the efficiency and the adaptability of an algorithm are guaranteed.

The invention relates to a system and a method for the optimizing control of the track of a rotary crane. The system comprises a crane optimization track control device, a movement mechanism position detection module and a position control module, wherein the crane optimization track control device is connected with the movement mechanism position detection module and the position control module, and the movement mechanism position detection module and a movement mechanism control module are connected with movement mechanisms. The method comprises the following steps of: pre-designing and optimizing coordinate movement tracks in a natural space; converting the coordinate movement tracks into coordinates of the movement mechanisms; and performing real-time control on each movement mechanism. The system and the method can conveniently realize track optimization and track tracing, effectively reduce the running distance, improve the movement speed, save the movement space, reduce shaking and twisting of objects, are convenient for positioning control, are convenient to realize lifting and synergistic operations of large-size objects by a plurality of cranes, realize safety, high efficiency, energy conservation and emission reduction in the lifting process, greatly reduce the labor intensity, and has good economical efficiency and wide application range.

The invention discloses an endoscope-holding robot for a minimally invasive abdominal operation. The endoscope-holding robot comprises a horizontal moving mechanism, a vertical moving mechanism, an endoscope gesture adjusting mechanism and an endoscope holding rack, wherein the horizontal moving mechanism is a dual-guide-rail and single-screw lead screw sliding platform which is arranged in a horizontal direction; the vertical moving mechanism is a dual-guide-rail and single-screw lead screw sliding platform which is arranged in a vertical direction, and the vertical moving mechanism is integrally installed on a sliding block of the horizontal moving mechanism and is driven to move left and right by virtue of the sliding block; the endoscope gesture adjusting mechanism comprises a link mechanism and a cross rotary base; an endoscope is clamped by a clamp on the endoscope holding rack; the endoscope is distributed in a vertical direction; and an extension line of the axis of the endoscope passes through the central point of the cross rotary base. The robot disclosed by the invention is simple in integral structure, compact and reasonable in layout, large in motion space, good in action coordination among various moving parts and flexible to operate and control; and especially, the robot has the characteristics of being high in precision and good in sensitivity and coordination when various actions are implemented during running.

Cross-environment rendering and user interaction support provide a seamless computing experience in a multi-operating system computing environment. The multi-operating system computing environment may include a mobile operating system and a desktop operating system running concurrently and independently on a mobile computing device. Full user interaction support is provided for redirected and/or mirrored applications that are rendered using an extended graphics context. An extended input queue handles input events from virtual input devices for remotely displayed applications. Remotely displayed applications are mapped to separate motion spaces within the input queue. The mobile computing device may be a smartphone running the Android mobile operating system and a full desktop Linux distribution on a modified Android kernel.

The invention discloses a rope-traction upper limb rehabilitation robot and a control method thereof, wherein, the rehabilitation robot comprises a central control device, a control mode select switch, a servomotor, hauling ropes, a supporting plate and a bracket, and can be divided into a bracket-type structure and an upright post-type structure. The control method comprises passive control and active control and can be switched between the two control ways. In the bracket-type structure, a motor M4 is connected with a main bracket and an auxiliary bracket, and the auxiliary bracket is controlled to rotate horizontally. Three cantilever beams of the auxiliary bracket are connected with three servomotors and can tug the supporting plate to move in three-dimensional space respectively through the ropes, and a forearm of a patient is put on the supporting plate to move in the three-dimensional space along with the supporting plate. In the upright post-type structure, the three servomotors fixed on an upright post can tug the supporting plate to move in three-dimensional space respectively through the ropes, and the forearm of the patient is put on the supporting plate to move along with the supporting plate. The rehabilitation robot can help the patient to tug an upper limb, is convenient for the patient to carry out rehabilitation training in larger motion space, and achieves optimized motor function restoration.

The invention discloses facial expression synthetic arithmetic based on a hybrid transformation strategy. Transition from motion capture data space to target face model motion space is achieved by building a redirecting model of radial basis function facial expressions based on geodesic distance. On a stage of facial expression animation, the result of space transformation is utilized, and local motion of a peak is calculated through a weighted local transformation method based on adjacent feature points. Simultaneously, the whole situation displacement of the peak is calculated through a whole situation transformation method based on radial basis function interpolation. In the end, the final displacement of the peak is obtained by mixing a local displacement and the whole situation displacement. According to the facial expression synthetic arithmetic based on the hybrid transformation strategy, the same catching order can be applied to different face models, model transformation is easy, and simultaneously, different catching orders can be used on the same target model, reusing of motion capture data is achieved, and simultaneously strong animation third dimension is achieved. The algorithm flow chart of the facial expression synthetic arithmetic based on the hybrid transformation strategy is shown in an attached map 1.

The invention discloses a three-dimensional translational parallel mechanism based on motion decoupling with the purposes of overcoming problems of a parallel mechanism including difficulty in calculation of forward kinematics, small motion space, strong coupling between the position and the direction of a movable platform (II), low response speed and high cost. The three-dimensional translational parallel mechanism based on motion decoupling comprises a fixed platform (I), the movable platform (II), a first branched chain (III) of the parallel mechanism, a second branched chain (IV) of the parallel mechanism, a third branched chain (V) of the parallel mechanism. The movable platform (II) is located above the fixed platform (I). The first branched chain (III) of the parallel mechanism, the second branched chain (IV) of the parallel mechanism and the third branched chain (V) of the parallel mechanism are located between the movable platform (II) and the fixed platform (I). Upper ends of the first branched chain (III) of the parallel mechanism, the second branched chain (IV) of the parallel mechanism and the third branched chain (V) of the parallel mechanism are fixedly connected with the movable platform (II). Lower ends of the first branched chain (III) of the parallel mechanism, the second branched chain (IV) of the parallel mechanism and the third branched chain (V) of the parallel mechanism are fixedly connected with the fixed platform (I).

The invention discloses an action recognition method T-STAM based on a double-flow space-time attention mechanism. The method comprises the following steps: S1, processing a video to obtain an opticalflow graph of an RGB frame; s2, fusing the channel attention network SE-Net into a double-flow basic network BN-Inception to obtain an SE-BN-Inception; s3, inputting the selected RGB frame and optical flow field information into SE-BN-Inception, and modeling the dependency relationship of different channels in the features to obtain a feature vector X of the video; s4, inputting the feature X into a CNN-based time attention network to calculate a time attention score corresponding to each frame; s5, inputting the feature X into a multi-space attention network, and extracting a plurality of motion space salient regions of the frame; s6, fusing the spatial and temporal features to further enhance the feature expression of the video, and fusing the two streams according to different weightsto obtain an action recognition result.

The invention discloses a cognitive radio and edge calculation method based on an industrial wireless network. According to the method, a cognitive radio and edge calculation model architecture basedon the industrial wireless network is erected firstly. Secondly, system models are established, including a network model, a service model and a calculation model. Afterwards, distributed random optimization is described, including system state space, motion space and reward and optimization targets. Finally, optimal distribution motion minimizing the online calculation complexity is sought by useof dynamic random optimization and an augmented Markov decision process (MDP), wherein a state transition probability is calculated firstly, then a system reward is updated, and finally the augmentedMDP is reconstructed and joint resource management is performed. Through experimental tests, compared with a conventional method, the method can obtain higher system benefits, provide higher throughput and reduce system transmission delay under different spectrum arrival rates and numbers of CRECs.

The invention discloses a six-degree-of-freedom underwater robot. The underwater robot comprises a casing, wherein a main posture adjustment advancing set which comprises three posture adjusting devices, a main dynamic advancing set which comprises two posture adjusting devices and an auxiliary posture adjusting set which comprises a posture adjusting device are arranged on the casing; each posture adjusting device comprises a flow guiding cover, a motor base, a motor and screw propellers; three flow guiding covers of the main posture adjustment advancing set are arranged in an isosceles triangle, the center line of each of the flow guiding covers is parallel to a Z axis, a plane formed by the center lines of two of the three flow guiding covers is parallel to a Y axis, and the distances from the other flow guiding cover to the two flow guiding covers are equal; the center line of each of two flow guiding covers of the main dynamic advancing set is parallel to an X axis, and a plane formed by the center lines of the two flow guiding covers is parallel to the Y axis; the center line of the flow guiding cover of the auxiliary posture adjusting set is parallel to the Y axis. According to the underwater robot disclosed by the invention, six posture adjusting devices are adopted to advance the underwater robot, so that the six-degree-of-freedom motion of the underwater robot in a motion space is realized.

The invention discloses a gravity unloading device for a multi-joint spatial mechanism. A mounting platform is arranged on a rack, and two-dimensional follow-up platforms are arranged on the platform;each follow-up platform can move in the two horizontal directions, a constant-tension control unit is arranged on the platform and connected with a force transmitting mechanism through flexible slings, and tension is controlled by collecting and releasing the slings; the force transmitting mechanism is connected with the multi-joint spatial mechanism needing to be experimented, and force transmitted by the slings is distributed to the spatial mechanism; in the experimenting process, the follow-up platforms trace movement of the spatial mechanism, so that the slings are in a vertical state allthe time; and the force transmitting mechanism make following movement at the same time, so that the extension lines of the slings correspond to the joint barycenters of components through the spatial mechanism. According to the gravity unloading device, the slings keep vertical through a horizontal position servo trace method, the tension of the slings is controlled to keep the tension of the slings, the number of the slings is small, the moving space is large, and the gravity unloading precision can be measured.

Machine motion is directed using a graphical interface by establishing an icon in an image, displaying the icon in images having perspectives such that the icon is rendered in a corresponding relative position in the second images and moving the icons in the second images when the icon is moved in the first image, the movement being constrained along a line projected from a camera point associated with the first image and a target destination. Actions are defined using the icon and a set of action descriptors. The dominant motion space is constrained to a spherical coordinate frame centered on a vision locus and machine motions are mapped to coincide therewith based on a graphical interface view.

The invention discloses a redundant flywheel set angular momentum self-management method. The method includes the steps of setting an unloading period T through periodic rules of three-shaft stable satellite external disturbance torque, estimating angular momentum HI(T) accumulated by the external disturbance torque in the unloading period, setting a composite angular momentum initial value H0 according to the HI(T), determining a working wheel set angular momentum target value h(t) and target angular momentum hZ of working wheel set zero movement space according to an initial working wheel set or a working wheel set where failure processing and reconstitution are conducted, determining working wheel set unloading target angular momentum h(0) according to hZ and H0, controlling the working wheel set angular momentum to be the unloading target angular momentum h(0) by controlling air injection and the wheel rotating speed at the unloading moment, and controlling the working wheel set angular momentum to be the working wheel set angular momentum target value h(t) according to a zero movement control law at the non-unloading moment. According to the redundant flywheel set angular momentum self-management method, unloading can be conducted in the optimized direction with the largest flywheel set angular momentum capacity, fuel is saved, and no air injection control is needed when a flywheel set is reconfigured in the non-unloading period.