124 results about "Gait control" patented technology

The invention discloses a gait control method and a device for a dual-foot robot. The method comprises the steps of selecting gait control parameters for a starting period, a walking period and a stopping period; acquiring the motion trajectory of the mass center of the dual-foot robot during the walking period when the zero moment point of the dual-foot robot is within a stable area; obtaining the first value of each gait control parameter when the mass center is at the initial moment of the walking period and the second value of the gait control parameter when the mass center is at the ending moment of the walking period; setting a first constraint condition at the ending moment of the starting period by utilizing the first value, and setting a second constraint condition at the starting moment of the stopping period by utilizing the second value; based on the first constraint condition and the second constraint condition, calculating the motion trajectory of the mass center during the starting period and during the stopping period respectively; controlling the walking of the dual-foot robot to enable the motion trajectory of the mass center of the dual-foot robot during walking to meet the motion trajectory of the mass center during the starting period, the walking period and the stopping period respectively. According to the technical scheme of the invention, the stable cohesion of the dual-foot robot during the starting period, the walking period and the stopping period and the walking stability of the dual-foot robot can be ensured.

The invention discloses a robot single-leg assembly control development performance test platform and method. The platform comprises a gantry three-coordinate mechanical arm assembly, a robot leg connection bracket, a Stewart platform, a six-dimensional force sensor, a robot single-leg assembly and a five-dimensional force measurement platform, wherein a servo controller and a displacement sensor are integrated in the Stewart platform; the Stewart platform is invertedly arranged on a base of the robot leg bracket; the five-dimensional force measurement platform is arranged at the center of the ground below the robot single-leg assembly; and the robot leg connection bracket is fixed on a Z-axis-direction mobile supporting frame assembly of the gantry three-coordinate mechanical arm assembly. The test platform is suitable for single-leg movement and quick gait control during bionic gait generation of a four-foot or multi-foot hydraulic driving robot as well as development and research on multiple control strategies for robot load distribution, control force distribution, single-leg force feedback control and 'discrete gait and continuous force control' attitude stabilization control.

The invention discloses a humanoid robot gait control method based on model correlated reinforcement learning. The method comprises steps of 1) defining a reinforcement learning framework for a stable control task in forward and backward movements of a humanoid robot; 2) carrying out gait control of the humanoid robot with a model correlated reinforcement learning method based on the sparse online Gaussian process; and 3) improving a motion selection method of a reinforcement learning humanoid robot controller by a PID controller, and taking the improved operation as an optimizing initial point for the PID controller obtaining the motion selection operation of the reinforcement learning controller. The invention utilizes reinforcement learning to control gaits of the humanoid robot in movement, and thus the movement control of the humanoid robot can be automatically adjusted via interaction with the environment, a better control effect is achieved, and the humanoid robot is enabled to be stable in forward and backward directions.

The invention provides a control system and method for learning variable impedance. The system mainly comprises a variable impedance controller, a system Gaussian process model, a variable impedance control strategy, and a strategy learning algorithm. The system does not need any priori knowledge of an environment, and the system Gaussian process model is constructed according to the interaction data. The long-term inference and planning of the system is carried out in a Bayesian mode. The system can extract more useful information from the limited observation data, and completes a complex force control task through the least interaction time. Through the adding of an energy los item to a cost function, the system achieves the balance between an error and energy, and enables a robot to begood in compliant capability. Finally, the obtained variable impedance control strategy allows the target rigidness and damping parameters to be adjusted at the same time at different stages of the task. The system and method can be widely used for the compliant control tasks: double-mechanical-arm assembly, multi-mechanical-arm cooperation and robot gait control, and guarantee the safety and robustness of the interaction operation.

Method and system for gait controlled operations of a device including one or more gait sensors (22) for sensing a pattern of self-propelled movement of a person, which defines a gait. The gait controlled device (10) includes at least one data processor (40) configured for determining one or more first parameters which define a first gait of an authenticated person based on a first output of the one or more gait sensors. The data processor uses a second output of the one or more gait sensors (22) to determine one or more second parameters which define a second gait of a second person. Thereafter, the data processor compares the first parameters to the second parameters to determine if the second gait resulted from the authenticated person. The data processor is configured to perform at least one action responsive to the comparing.

The invention discloses a gait control method and device of a multi-legged robot and the robot. The method comprises the following steps: each gait of the robot is divided into multiple action units with a preset sequence, and leg end motion trajectories of the robot and target zero torque points corresponding to the leg end motion trajectories are obtained; under the condition of the robot walking according to preset gaits, a mass center motion trajectory of the robot is obtained according to present zero torque points and target zero torque points of the robot obtained in real time; trajectory information of part connecting nodes of the robot is determined according to the leg end motion trajectories and the mass center motion trajectory; and the robot is controlled to move according tothe sequence of the action units in the preset gaits and the trajectory information of the part connecting nodes corresponding to the action units. The method solves the technical problem of weak interference resistance caused by offline design in operation planning of a double-legged robot in the prior art.

The invention discloses an optimal control method of a gait of a humanoid robot based on a deep Q network. The method comprises the steps of: constructing a gait model, and obtaining interaction databetween the humanoid robot and the environment during walking for providing training samples; performing learning and training on the deep Q network based on the training samples of a memory data poolto obtain a state-action strategy deep Q network model of the humanoid robot; obtaining state parameters of the humanoid robot in an action environment to serve as the input of the deep Q network model, and obtaining action parameters of the deep Q network model under the current state-action strategy; performing gait control on the humanoid robot by using the constructed gait model and accordingto the action parameters output by the deep Q network model; and achieving the purpose of updating the deep Q network in the training of the deep Q network model by generating an award function. By adopting the optimal control method disclosed by the invention, the walking speed of the humanoid robot can be improved, and the fast and stable walking of the humanoid robot can be realized.

The invention discloses a method for controlling the time of foot landing which can lead a humanoid robot walk steadily and a system thereof, belonging to the field of robots. The method comprises that: the feet of the swing legs of a robot are detected to judge whether the feet land or not; if the feet of the swing legs of the robot land before the time of the programming pace, the ankle of the swing legs is elevated; if the feet of the swing legs of the robot land after the time of the programming pace, the ankle of the swing legs is lowered. The system comprises a detection device and an adjustment device; wherein, the adjustment device comprises an advanced adjustment module, a lagging adjustment module and a supporting adjustment module. The invention has the effect that the robot can walk steadily in the actual environment by amending the programming dynamic pace of the robot and controlling the pace in real time.

A method and a device for controlling a gait of a biped robot. The method includes: selecting gait controlling parameters, and acquiring a movement trajectory of a center of mass when a zero moment point of the biped robot is located within a steady area; obtaining first numerical values of each of the gait controlling parameters of the center of mass and second numerical values of the center of mass; setting a first constraint condition when the step starting phase ends by using the first numerical values, and setting a second constraint condition when the step ending phase starts by using the second numerical values; calculating the movement trajectories of the center of mass in the step starting phase and the step ending phase on the basis of the first constraint condition and the second constraint condition, respectively; and controlling a walking of the biped robot.

The invention belongs to the field of medical apparatus, and discloses an exoskeleton coordinated gait control method based on human gait movement coordination characteristics. The exoskeleton coordinated gait control method includes the steps that (a) for a lower limb exoskeleton to be controlled, the respective joint angle and angular velocity of a hip joint and a knee joint of the lower limb exoskeleton are detected in real time, detection results are compared with predetermined swing phase switching conditions, and a first state satisfying the swing phase switching conditions is taken as astarting point of a swing phase; (b) an ideal coordinated movement curve between the angles of the hip joint and the knee joint under the swinging phase is constructed; and (c) an angle theta of thehip joint and the knee joint of the lower limb exoskeleton to be controlled is monitored in real time at each moment after entering into the swing phase, the theta is compared with the ideal coordinated movement curve, when the acceptable conditions are not met, the lower limb exoskeleton applies torque on a man-machine system, and in this way, control of coordinated gait of the lower limb exoskeleton is realized; and through the exoskeleton coordinated gait control method, the control of the coordinated gait of the walking exoskeleton is realized, and at the same time movement autonomy of a user is maintained.

The invention discloses a biped robot gait control method and a biped robot. The method comprises the steps of obtaining six-dimensional force information, and judging a supporting state of the bipedrobot according to the six-dimensional force information; calculating the ZMP position of each foot of the biped robot according to the six-dimensional force information; setting a ZMP expected valueof each foot according to the supporting state and a body posture error; determining an ankle joint compensation angle of each foot of the biped robot according to the ZMP position, the change rate ofthe ZMP position, the ZMP expected value and the change rate of the ZMP expected value; tracking and adjusting the current ankle joint angle of the biped robot in real time according to the ankle joint compensation angle; and circulating the steps at a set frequency until posture control or ankle joint compliance is completed in different supporting states.

In a control system of a legged mobile robot having a body and legs connected to the body and driven by a leg actuator, there is provided an operation controller which generates gaits based on an external force, more specifically gaits for walking by taking a hand of a human being or with the hand being taken by the hand of the human being and controls operation of at least the leg actuator based on the generated gaits. With this, it becomes possible to control the robot to come contact with a human being to establish communication therewith, while enabling to keep a stable posture during the contact.

The invention discloses a multi-motion mode moving parallel mechanism suitable for unstructured terrain and a gait control method. The multi-motion mode moving parallel mechanism suitable for the unstructured terrain comprise an upper platform, a lower platform, Hooke hinge assemblies and connecting rods, wherein the upper platform and the lower platform are arranged oppositely in parallel in a spaced mode, the ends of the upper platform and the lower platform are provided with the Hooke hinge assemblies and rotatably connected with the Hooke hinge assemblies, steering engines are installed atthe ends, far away from the upper platform and the lower platform, of the Hooke hinge assemblies, and the connecting rods are installed between the Hooke hinge assemblies at the same ends of the upper platform and the lower platform. The gait control method comprises a full-posture rolling mode, a variable-width rolling mode, a switching mode and a side rolling mode. The multi-motion mode movingparallel mechanism suitable for the unstructured terrain and the gait control method have the beneficial effects that the adaptability is good, and the application prospects are very good; and the parallel mechanism is adopted overall, and compared with a traditional chain opening mechanism, the parallel mechanism has the advantages of being simple in structure, good in rigidity and high in bearing capacity. According to the foldable multi-motion mode moving parallel mechanism, the steering engines in the mechanism can be controlled in a time-sharing manner, and free switching among differentmodes can be realized.

The invention provides a single lower limb rehabilitation exoskeleton device and a control method thereof. The device comprises a controller, and a healthy side lower limb assembly and an injured sidelower limb assembly in communication connection with the controller; the controller is used for determining the current state of an uninjured side lower limb through the healthy side lower limb assembly and determining the current state of an injured side lower limb through the injured side lower limb assembly; the uninjured side lower limb assembly is used for collecting motion data of the healthy side lower limb when the healthy side lower limb is in a lifting state and sending the motion data to the controller; the controller is used for determining gait data corresponding to the injured side lower limb assembly according to the motion data of the healthy side lower limb and sending the gait data to the injured side lower limb assembly; and the injured side lower limb assembly is usedfor driving the injured side lower limb to move according to the gait data when the healthy side lower limb is in the supporting state. According to the single lower limb rehabilitation exoskeleton device and the control method thereof, autonomous gait control of a patient can be well realized, and information interaction between the patient and the single lower limb rehabilitation exoskeleton device can also be realized, so that personalized physical rehabilitation training is realized.

The embodiment of the invention discloses a humanoid gait control method, device and equipment of a humanoid robot and a storage medium, and the method comprises the steps: obtaining a first vector from a virtual centroid to a left ankle joint and a second vector from the virtual centroid to a right ankle joint of the humanoid robot at a current moment, and obtaining a virtual centroid original planning value of the humanoid robot at the current moment; according to the first vector, the second vector, the virtual centroid original planning value and a preset algorithm for reducing the virtual centroid height, determining the target virtual centroid height of the humanoid robot after the virtual centroid is reduced at the current moment; and controlling the humanoid robot to walk straight knees according to the target virtual centroid height. The height of the virtual centroid of the humanoid robot is reduced by using the vector between the two legs and the virtual centroid of the humanoid robot, straight knee walking is realized by reducing the height of the virtual centroid, gait control of the humanoid robot is optimized, the torque requirement of a knee joint steering engine is further reduced, and the hardware design weight is reduced.

The invention relates to a snake imitation search and rescue robot multi-gait control method, comprising: according to natural, wheel type and chaotic motion gaits, inputting a consistent relation by utilizing the output of a Hopfiled Neural Network (HNN) and the need of three types of motion gaits; adjusting the weighting parameter of the HNN to obtain a corresponding gait signal; using the output signal of the HNN as the input signal of a snake imitation search and rescue robot dynamics controller; and controlling the snake imitation search and rescue robot to generate a corresponding gait. In a non stable state, i.e., at an initial time and a gait conversion time, the motion shape code technique (MSCT) is utilized to generate the initial shape of a corresponding gait, and a linear shape is used as an intermediate process state to perform gait conversion. A traditional central pattern generator (CPG) method does not consider non period gait and chaotic gait problems in gait control, while the snake imitation search and rescue robot multi-gait control method solves the problem, is more practical, and allows a snake imitation search and rescue robot to be better applied to search and rescue tasks.

The invention discloses a gait training method and device of a quadruped robot based on deep reinforcement learning, electronic equipment and a medium. According to the application, an environment, the quadruped robot model and a sensor model can be built in the simulator; a measurement signal is acquired through a multi-source information fusion technology and a sensor, and the state of the robotis estimated; a foot end track planning and initial gait control strategy of the quadruped robot is generated by utilizing the environment module, the robot model and the measurement signal; and strategy improvement is performed on the initial gait control strategy of the quadruped robot based on a deep reinforcement learning algorithm until a strategy improvement ending condition is triggered. By applying the technical scheme of the invention, the problems that a large amount of manual parameter adjustment is needed and the labor cost is relatively high due to a model control algorithm in related technologies can be avoided. Parameters obtained through training are often poor in robustness, can only be suitable for a certain fixed scene, and are difficult to meet the requirements of various application tasks of an actual quadruped robot.

The invention discloses a humanoid intelligent robot system. The humanoid intelligent robot system is characterized in that the system comprises a robot body and a control system; the control system comprises a visual image acquiring module, a central processing module, a drive module and a sensing module; and a pre-planed real-time gait control algorithm and a binocular vision based obstacle avoidance algorithm are established in the central processing module to control motion of a robot. The gait of the humanoid robot can be accurately controlled through combined usage of the two algorithms,so that the humanoid robot better achieves the functions of obstacle avoidance and gait adjustment and the adaptability of the humanoid robot to a pavement is improved.

The invention relates to the field of fitness equipment, in particular to a gait regulation system of a treadmill, include a bearing plate, a mounting frame, a lifting mechanism, a mounting plate, a slope adjusting mechanism and a gait adjusting mechanism are provided with a running support assembly in the mounting frame, two connecting plates arranged at intervals are arranged at the top of the right end of the bearing plate, a rotating shaft is arranged between the two connecting plates, the gait adjusting mechanism comprises a finite position assembly and a sensing assembly, and the bearingplate is provided with a through slot through which the limit assembly passes. The invention has the beneficial effect that the gait adjusting mechanism can make the user have the same gait when running, the invention can meet the needs of the movement step of different users, the angle between the mounting frame and the horizontal plane can be formed automatically by the gradient adjusting mechanism, and the height of the armrest can be adjusted automatically by the lifting mechanism, so that the arm can be comfortably placed on the armrest by each person of different height.

The invention discloses a buffering / walking integrated hexapod lander and a gait control method thereof, belongs to the technical field of deep space star catalogue detection mechanisms, is composed of a mechanical structure, a control system and gait planning, and has two functions of stable landing and flexible roaming. The mechanical structure is composed of a hexagonal frustum type body and six buffering / driving integrated three-degree-of-freedom leg and foot mechanisms which are symmetrically distributed according to the center of a regular hexagon. The control system consists of a main control module and a special motion controller and is used for controlling the position of each driving motor in each buffering / driving integrated three-degree-of-freedom leg-foot mechanism so as to realize quasi-static walking of the lander; the gait planning adopts a stable and efficient 3-3 gait which is more matched with the hexapod configuration, and the smooth fluctuation track of the centroid of the body is realized through the motion control of each buffer / drive integrated three-degree-of-freedom leg and foot mechanism.

The invention relates to a hydraulic hexapod robot and a walking gait control method based on mass center fluctuation, which belong to the technical field of robots. The robot comprises a body and three hydraulic mechanical legs arranged on the two sides of the body respectively. Each hydraulic mechanical leg comprises a root joint, a hip joint, a thigh rod, a knee joint, a shank rod and a foot end fixedly arranged on the shank rod, and the gait control method comprises the steps that the hydraulic mechanical legs are controlled to walk according to the three-foot gait according to the plannedfoot end track, so that the mass center track of the six-foot walking robot is a cosine curve track; specifically, under a calcaneal joint coordinate system, the components of the foot end track in the advancing direction and the vertical direction are all sextic polynomials, and displacement, speed and acceleration are continuous. Compared with a control method for keeping the mass center heightapproximately unchanged in the prior art, the control method provided by the invention has the advantages that the average flow and average power of the robot in one gait period can be effectively reduced through the gait with the fluctuated mass center, so that the energy utilization rate is effectively increased, and the method can be widely applied to the technical field of robots.

The invention discloses an external-field-driven double-leg standing and walking magnetic micro-robot and a gait control method thereof, and relates to the field of micro-robots and control thereof. The invention aims to solve the problems of single motion form and complex driving mode of a conventional micro-robot on the solid surface. The micro-robot disclosed by the invention is of a sheet-shaped structure with two legs. According to the gait control method of the micro-robot, three pairs of orthogonal Helmholtz coils are used for generating an oscillation magnetic field, and the micro-robot is driven to stand and walk on the solid surface. The coils generate a constant magnetic field in the vertical direction to enable the micro-robot to stand, and a swing magnetic field in the horizontal direction drives the micro-robot to walk. The walking direction of the micro-robot can be changed by adjusting the swing direction of the horizontal magnetic field, and the advancing speed of themicro-robot can be changed by changing the swing amplitude and frequency of the magnetic field. The method is suitable for the fields of medical treatment, microoperation and the like.

The invention provides a snakelike robot adaptive path following controller based on tracking error and time-varying coefficient prediction and a design method thereof. Based on a multi-joint snakelike robot kinematics model and a control function of gait joints, according to a dynamic control task and a posture control task in a snakelike robot path following process, based on the predicted value of the motion position error of the snakelike robot, the predicted value of the time variable of the sideslip angle and the predicted value component of the interference variable, the expected motion direction angle of the snakelike robot is obtained through an improved LOS method, and gait control over the multi-joint snakelike robot is carried out; and according to the virtual auxiliary function of the motion direction angular velocity of the multi-joint snake-shaped robot and the joint angle compensation function of the snake-shaped robot, based on an updating law of a parameter predicted value, asymptotic stability of a motion direction angle error, a joint angle error and a predicted value error of the multi-joint snake-shaped robot is achieved. According to the method, the robot can predict tracking position errors and interference variables.

The invention relates to an improvement on the robot technology, in particular to a humanoid robot gait planning deep reinforcement learning new method. The problem of walking control of a biped robotis solved through an improvement DQN algorithm. On the basis that there is no need to establish a complex biped robot dynamic model, the deep reinforcement learning method is used for realizing long-distance stable gait control of the biped robot under the condition of walking fast. The method comprises the steps of establishing of a humanoid biped robot model, pre-training control parameters andwalking movement of the biped robot, wherein a simplified six-degree-of-freedom connecting rod model is adopted as the humanoid biped robot model, the feet adopt planar sole structures; through the pre-training control parameters, training of a controller is carried out through the improved deep reinforcement learning method adopting a DQN network structure; and the stability of walking movementof the biped robot is realized through the trained controller.

The invention provides an exoskeleton robot, a walking aiding control method, a terminal and computer equipment. The exoskeleton robot comprises at least one moving joint, an obtaining unit and a control unit, wherein the at least one moving joint is provided with a motor, the obtaining unit is used for obtaining the motion information of the at least one moving joint, and the control unit is usedfor generating the corresponding gait data according to the motion information, sending the gait data to the terminal through a communication unit to make the terminal generate a corresponding gait control curve according to the gait data and controlling the motor to drive the at least one moving joint according to the gait control curve. By means of the technical scheme of the exoskeleton robot,the walking aiding control method, the terminal and the computer equipment provided by the invention, a complex and expensive sensor control system does not need to be arranged on an exoskeleton robot body, the hardware cost of the exoskeleton robot is reduced, by sending the gait data to the terminal, health care professionals are made to be capable of timely grasping the gait situation, and then a targeted treatment, walking aiding or gait correction scheme is specified, and the interaction convenience of the exoskeleton robot is improved.

The invention provides an eight-legged robot with bionic rigid-flexible coupling legs and a control method. The search and rescue robot comprises a machine frame, a machine shell, the rigid-flexible coupling legs, a control processing module, a sensing unit, a wireless transmission module, a motor driving module and a power module, eight rigid-flexible coupling legs with a same structure are arranged, a first left leg, a second left leg, a third left leg and a fourth left leg are installed on the left side of the machine frame, and a first right leg, a second right leg, a third right leg and a fourth right leg are installed on the right side of the machine frame and each comprises a rigid segment, a joint shaft, a foot, a torsion spring, a linear driver, a connecting piece and a spring steel sheet. The control method comprises a step length control method, a step frequency control method, a height control method, a rack and rigid-flexible coupling leg coordination control method, a stable climbing gait control method and a tunnel passing gait control method. According to the eight-legged robot, the rigid segments and the flexible spring steel sheets are combined, and the eight-legged robot has the advantages of being compact in structure, light, self-adaptive, flexible and the like and can be used for tasks such as search and rescue and detection.

The invention belongs to field of robot walking control, and in particular relates to a delayed feedback-based biped robot walking non-monocyclic gait control method which is characterized by determining a perturbation parameter p among robot walking control parameters; determining a Poincare section, a target fixed point and a nominal parameter corresponding to target fixed point; building a linearized model at a monocyclic fixed point; determining the feedback gain Kf and feedback gain Kp of a delayed feedback controller; determining the limiting value of the maximum perturbation amount; setting an original state and control parameters of a robot, solving a perturbation parameter pi of a former step, and if the perturbation parameter pi does not exceed the limit of the maximum perturbation amount, executing the perturbation parameter; if the perturbation parameter pi exceeds the limit of the maximum perturbation amount, executing the nominal parameter p*; and when a swinging leg falls to the ground, stopping the current step, and controlling a next step. According to the delayed feedback-based biped robot walking non-monocyclic gait control method, the gait can be restrained step by step through parameter perturbation by adopting only one group of constant state feedback gain, an expected fixed point can be reached fast and stably, the non-monocyclic gait is realized, and the robot walking parameter range is enlarged.

The invention discloses a control method of a bionic snake-shaped robot. A CPG mathematical model of the bionic snake-shaped robot is obtained through an electromechanical system in combination with simulation firstly. For the control over the joint gait of the robot and the mode switchover of the gaits, periodic oscillation signals generated by the CPG mathematical model are used for controlling the rhythmic movement of the robot. Iterative refinement is carried out on the unilateral and bilateral link weights in a CPG network through a genetic algorithm (GA). The feedback about the direction and speed of a motor is carried out in real time through a potentiometer, and obtained signals are used for carrying out feedback adjustment on output of feedback control and CPG control. Traditional dynamics computation and modeling are avoided, errors are reduced, multiple gaits and multiple phase relationships of the robot can be realized, and gait switchover can be directly realized in computer software. Instantaneity and controllability are improved, the bionic requirements are better met, and the multiple gaits and the function of smooth switchover among the gaits of the snake-shaped robot are realized.

The invention discloses a lower limb rehabilitation training machine which comprises a mechanical structure unit and a control unit. The mechanical structure unit comprises a base support, a gait control motor, a driving sprocket installed at the output end of the gait control motor and a driven sprocket in transmission connection with the driving sprocket through a transmission chain. Cranks are installed on the two sides of the driven sprocket respectively. A connecting rod is hinged to the end, away from the axis of the driven sprocket, of each of the two crank. A movable roller is installed at the end, away from the crank, of each connecting rod. A pedal plate is installed on each connecting rod. The control unit comprises a single chip microcomputer and a peripheral circuit. The signal output end of the control unit is connected to the gait control motor through a speed controller. The training machine is used for promoting lower limb rehabilitation of a user, speed control over the motor is performed through the single-chip microcomputer, the pace frequency can be regulated according to rehabilitation requirements of the user at different stages, and the lower limb rehabilitation training machine is reliable in whole structure, stable in operation, good in simulation ability, and suitable for being applied and popularized in lower limb rehabilitation.