摘要:
A legged mobile robot, a legged mobile robot controller and a legged mobile robot control method are provided to perform a loading operation to load a gripped object in parallel on a target place having a height where a stretchable range of arm portions of the legged mobile robot is enhanced with no operator's handling. The legged mobile robot includes the arm portions having links for gripping an object, and leg portions having links for moving, and the arm and the leg portions are joined to a body thereof. The legged mobile robot controller includes a data acquisition unit, a whole-body cooperative motion control unit and a loading detection unit, and controls motions of the legged mobile robot based on posture/position data regarding a posture/position of each link of the legged mobile robot and on an external force data regarding an external force affecting the arm portions.
摘要:
A legged mobile robot, a legged mobile robot controller and a legged mobile robot control method are provided to perform a loading operation to load a gripped object in parallel on a target place having a height where a stretchable range of arm portions of the legged mobile robot is enhanced with no operator's handling. The legged mobile robot includes the arm portions having links for gripping an object, and leg portions having links for moving, and the arm and the leg portions are joined to a body thereof. The legged mobile robot controller includes a data acquisition unit, a whole-body cooperative motion control unit and a loading detection unit, and controls motions of the legged mobile robot based on posture/position data regarding a posture/position of each link of the legged mobile robot and on an external force data regarding an external force affecting the arm portions.
摘要:
A power device 1 has a speed reducer 3 and a spring member 4 provided in a power transmission system between an actuator 2 and a driven element 5, an operation target of the actuator 2 is determined according to a linear combination of values that are obtained by passing through low-pass filters, a deviation Δω12 between an estimated value ω1—e of a velocity of an input unit 4a of the spring member 4 and an estimated value ω2—e of a velocity of the driven element 5, and a deviation Δθdef between a value θdef—s of a deviation between the input unit 4a of the spring member 4 and the driven element 5, and a target value θdef_cmd of the displacement difference corresponding to a target driving force τref of the driven element 5.
摘要:
An element 22 which determines a manipulation amount for controlling a motion state of each joint 4 of a robot 1 calculates a pseudo inverse matrix A* used for calculating the manipulation amount, using a value of an adjustment parameter k determined so that an absolute value of a determinant DET is equal to or more than a predetermined threshold. Setting a provisional value of the adjustment parameter k by gradually increasing the provisional value from a predetermined initial value, calculating the determinant DET using the set provisional value, and determining whether the absolute value of the calculated determinant DET is equal to or more than the predetermined threshold is repeated, and the provisional value of the adjustment parameter k when the determination result is true is determined as the value of the adjustment parameter k used for the calculation of the pseudo inverse matrix A*.
摘要:
Disclosed is a robot control device including a means for determining control inputs classified by state quantity for achieving respective target values of a plurality of types of state quantities of a robot, and a means for determining a synthesized control input by synthesizing control inputs classified by frequency region while determining control inputs classified by frequency region in a plurality of respective frequency regions, according to control inputs classified by state quantity. The means determines a control input classified by frequency region corresponding to any one of the frequency regions by synthesizing the plurality of control inputs classified by state quantity in a mutually non-interfering manner. The operation of the robot is controlled so that, under a variety of operating conditions of the robot, a plurality of types of state quantities are efficiently controlled to target values which correspond to the respective types of state quantities.
摘要:
In a power device 1 having a speed reducer 3 and a spring member 4 having elastic characteristics to a power transmission system between an actuator 2 and a driven element 5, an operation target of the actuator 2 is determined according to a linear coupling value of values that are obtained by passing through low-pass filters, a deviation Δω12 between an estimated value ω1_e of a displacement velocity of an input unit 4a of the spring member 4 and an estimated value ω2_e of a displacement velocity of the driven element 5, and a deviation Δθdef between a measured value θdef_s of a deviation difference between the input unit 4a of the spring member 4 and the driven element 5, and a target value θdef_cmd of the displacement difference corresponding to a target driving force τref of the driven element 5.
摘要:
The robot control apparatus has an input-output section, a control section including a priority data generation section, a schedule data generation section, an execution command generation section, and a task data dividing section, a map information database, individual information database, a robot information database, and a task information database. The priority data generation section generates priority data for task data, stored in a task control database that that has not been executed. The schedule data generation section generates schedule data by assigning tasks to the robots on the basis of the priority data to generate schedule data. The execution command generation section generates execution commands for causing the robots to execute the tasks.
摘要:
A parking lot attendant robot is provided with a camera for capturing an image of an vehicle and a parking space, and a receiver from receiving vehicle information such as the steering angle, shift position and vehicle speed from the vehicle. Based on the knowledge of the position of the vehicle with respect to the parking space gained from the camera and the vehicle information received from the vehicle, the robot computes the optimum parking path which the vehicle should follow. The robot signals the motorist by a visual and/or auditory display and/or a hand signal as to how the steering wheel should be turned and which way the vehicle should move or stop for the vehicle to follow this parking path and to be successfully parked in the prescribed parking space.
摘要:
Disclosed is a robot control device including a means for determining control inputs classified by state quantity for achieving respective target values of a plurality of types of state quantities of a robot, and a means for determining a synthesized control input by synthesizing control inputs classified by frequency region while determining control inputs classified by frequency region in a plurality of respective frequency regions, according to control inputs classified by state quantity. The means determines a control input classified by frequency region corresponding to any one of the frequency regions by synthesizing the plurality of control inputs classified by state quantity in a mutually non-interfering manner. The operation of the robot is controlled so that, under a variety of operating conditions of the robot, a plurality of types of state quantities are efficiently controlled to target values which correspond to the respective types of state quantities.
摘要:
A control device for a mobile robot, in which the desired value of a motion state amount of a mobile robot includes at least the desired value of a vertical component of a first-order differential value of the translational momentum of the entire mobile robot. The desired value is determined by a state amount desired value determiner such that the observed value of the vertical position of an overall center-of-gravity point of the mobile robot is converged to a predetermined desired value according to a feedback control law. A control input determiner carries out the processing of inverse dynamics calculation, using the desired value of the motion state amount thereby to determine the desired driving force for each joint. The operation of an actuator is controlled on the basis of the determined desired driving force.