Techniques are described herein for generating a dynamical systems control policy. A non-parametric family of smooth maps is defined on which vector-field learning problems can be formulated and solved using convex optimization. In some implementations, techniques described herein address the problem of generating contracting vector fields for certifying stability of the dynamical systems arising in robotics applications, e.g., designing stable movement primitives. These learning problems may utilize a set of demonstration trajectories, one or more desired equilibria (e.g., a target point), and once or more statistics including at least an average velocity and average duration of the set of demonstration trajectories. The learned contracting vector fields may induce a contraction tube around a targeted trajectory for an end effector of the robot. In some implementations, the disclosed framework may use curl-free vector-valued Reproducing Kernel Hilbert Spaces.

A method for controlling a robotic device. The method includes providing demonstrations for carrying out a skill by the robot, each demonstration including a robot pose, an acting force as well as an object pose for each point in time of a sequence of points in time, ascertaining an attractor demonstration for each demonstration, training a task-parameterized robot trajectory model for the skill based on the attractor trajectories and controlling the robotic device according to the task-parameterized robot trajectory model.

There is provided a method and computer program product for programming a robot by manually operating it in gravity-compensation kinesthetic-guidance mode. More specifically there is provided method and computer program product that uses kinesthetic teaching as a demonstration input modality and does not require the installation or use of any external sensing or data-capturing modules. It requires a single user demonstration to extract a representation of the program, and presents the user with a series of easily-controllable parameters that allow them to modify or constrain the parameters of the extracted program representation of the task.

A method of generating a control program for a robot includes generating a trajectory in which a robot arm moves between a plurality of teaching points based on a first constraint condition with respect to a movement time of the robot arm and a second constraint condition with respect to a drive condition for driving the robot arm by a processor, displaying the trajectory generated by the processor and accumulated power consumption when the robot arm moves along the trajectory by a display unit, and, when receiving an instruction to employ the trajectory, generating a control program for the robot based on the trajectory by the processor.

The articulated-arm robot has a robot arm with an arm element movable via a joint and a sensor for continuously measuring a status parameter of the joint. The articulated-arm robot also has an optical signaling device arranged on the robot arm in spatial assignment to the joint and an assessment device for continuously assessing the measured status parameter in a joint-specific manner and for controlling the signaling device on the basis of the assessment result.

The invention concerns a robotic arm (1) with at least two arm modules (41, 42) which are moveable relative to one another and at least one manually operable input module (11) for generating control signals for the control of the robotic arm (1) on the basis of a user input. Both arm modules (41, 42) have a first interface (38, 40) onto which the input module (11) can be selectively mounted.

In teaching of a robot, a control device controls a movable unit in a first control mode in which the movable unit continuously moves according to a force detected by a force detector and a second control mode in which the movable unit moves by a predetermined movement amount according to the force detected by the force detector. A controller selects a first control mode or a second control mode according to a temporal change in the force detected by the force detector and a magnitude of the force.

According to an aspect of the present disclosure, there is provided a transfer system comprising a transfer robot configured to transfer a transfer target object by an end effector based on an operation instruction, and a controller configured to output the operation instruction to the transfer robot, wherein at least any one of the end effector and the transfer target object has at least any one of a sensor and a camera, the controller calculates a relative position between the end effector and the transfer target object based on at least any one of a detected result of the sensor and a captured result of the camera, and the controller determines a teaching position of the end effector with respect to the transfer target object based on the relative position, and outputs the operation instruction to the transfer robot such that the end effector is disposed at the teaching position.

A teaching method for a transfer device provided with a pick configured to hold a substrate and a mapping sensor, includes detecting a position of an edge of the substrate arranged in a teaching target module by the mapping sensor and setting a teaching position in one horizontal direction, transferring the substrate from the teaching target module to a stage of an alignment device by the pick based on the set teaching position, rotating the stage by a predetermined angle and detecting a locus of the position of the edge of the substrate, and estimating an eccentricity amount between the stage and the substrate based on the detected locus of the position of the edge of the substrate.

A method and apparatus for controlling a robot is provided. In this robot, direct teaching can be performed while updating a position command on the basis of an applied external force. In the method and apparatus, a proximity region is set inside a boundary of an operation-allowed range of the robot, the proximity region being indicative of a proximity of the boundary. Stored is an external force applied when a monitoring point provided in the robot reaches the proximity region as a reference external force. And performed is comparing the reference external force with a current external force when a current position of the monitoring point is in the proximity region, to thereby determine a direction that facilitates movement away from the proximity region.