A method for calculating an optimized trajectory with a simulation program and an optimization routine includes making available a trajectory and adjusting it to boundary conditions, implementing a loop having the steps of a provision of one first trajectory, a modification of a (further) trajectory and adjustment of the (further) trajectory on the basis of boundary conditions, using as the optimized trajectory a trajectory which has been made available on the basis of an extremal or predetermined parameter, and, after being calculated, making available the optimized trajectory to a control device in order to move a holder for a component part.

Various embodiments of the technology described herein generally relate to systems and methods for trajectory optimization with machine learning techniques. More specifically, certain embodiments relate to using neural networks to quickly predict optimized robotic arm trajectories in a variety of scenarios. Systems and methods described herein use deep neural networks to quickly predict optimized robotic arm trajectories according to certain constraints. Optimization, in accordance with some embodiments of the present technology, may include optimizing trajectory geometry and dynamics while satisfying a number of constraints, including staying collision-free, and minimizing the time it takes to complete the task.

A robot system including a first robot, a second robot, a first robot control device for controlling the first robot, a second robot control device for controlling the second robot, and a sensor, the first and second robots carrying out a coordinated operation, wherein: the first robot control device includes a first trajectory planning unit for planning the action of the first robot, a first control unit for executing the planned action of the first robot, a first collision prediction unit for predicting a collision between the first or second robot and an external disturbance on the basis of information from the sensor, and a first interlocking-preventing unit; the second robot control device includes a second trajectory planning unit for planning the action of the second robot, a second control unit for executing the planned action of the second robot, a second collision prediction unit for predicting a collision between the first or second robot and an external disturbance on the basis of information from the sensor, and a second interlocking-preventing unit; the first interlocking-preventing unit generates an external-disturbance-avoiding trajectory for the first robot and transmits the external-disturbance-avoiding trajectory to the second interlocking-preventing unit when the first collision prediction unit predicts a collision with the external disturbance; and the first and second robots carry out actions to avoid the external disturbance. This makes it possible to prevent the occurrence of faults if the action of a robot is suddenly changed due to an external disturbance or the like when a plurality of robots carry out a coordinated operation.

A mobile robot employing a method and a system for localizing the mobile robot using external surveillance cameras acquires images from the surveillance cameras installed indoors adjacent to each other, recognizes objects included in the images by removing shadows from the images and performing a homography scheme, and avoids the recognized objects. The mobile robot employing the method and the system for localizing the mobile robot using the external surveillance cameras enables rapider localization and lower price as compared with a conventional image-based autonomous robot, so that the commercialization of a service robot is accelerated.

A mobile robot employing a method and a system for localizing the mobile robot using external surveillance cameras acquires images from the surveillance cameras installed indoors adjacent to each other, recognizes objects included in the images by removing shadows from the images and performing a homography scheme, and avoids the recognized objects. The mobile robot employing the method and the system for localizing the mobile robot using the external surveillance cameras enables rapider localization and lower price as compared with a conventional image-based autonomous robot, so that the commercialization of a service robot is accelerated.

A processing system is disclosed for providing processing of objects that include a programmable motion device including an end effector, a perception system for recognizing any of the identity, location, and orientation of an object presented in a plurality of objects at an input location, a grasp acquisition system for acquiring the object using the end effector to permit the object to be moved from the plurality of objects to one of a plurality of destination bins, and a motion planning system for determining a changing portion of a trajectory path of the end effector from the object to a base location proximate to the input location, and

A simulation system to determine an optimal trajectory path for a robot with an attached implement includes a trajectory simulator which provides a simulated trajectory path for an implement, an implement model database which comprises motion data of the implement, and a logger that associates a time stamp of the implement's motion during the simulated trajectory path to generate logger data. A profile is determined by the logger data received from the logger which identifies implement motion that exceeds predetermined thresholds, and a tuner adjusts the simulated trajectory path so as to reduce the number of times predetermined thresholds are exceeded.

A method substantially simultaneously plans a path and maps an environment by a robot. The method determines a mean of an occupancy level for a location in a map. The method also includes determining a probability distribution function (PDF) of the occupancy level. The method further includes calculating a cost function based on the PDF. Finally, the method includes simultaneously planning the path and mapping the environment based on the cost function.

A method for training a control strategy. The method includes providing training data, which demonstrate a control behavior, according to which control actions are to be generated, and training the control strategy with the aid of imitation learning by minimizing a measure of deviation between the distribution of state transitions according to the control strategy and the distribution of state transitions according to the demonstrated control behavior using the training data.

A robot control device includes: a reliability computing unit that is inputted with a feature quantity obtained from a sensor signal indicating a measurement result obtained by an external sensor installed in a main body of a robot or a surrounding environment of the robot, and computes a reliability for the sensor signal on the basis of a temporal change or a spatial change of the feature quantity; a correction command value computing unit that computes a trajectory correction amount for correcting a trajectory of the robot on the basis of the reliability and correction information calculated on the basis of the feature quantity; and a command value generation unit that generates a location command value for the robot on the basis of a predetermined target trajectory of the robot and the trajectory correction amount.