Systems, methods, devices, and other techniques are described for planning motions of one or more robots to perform at least one specified task. In some implementations, a task to execute with a robotic system using a tool is identified. A partially constrained pose is identified for the tool that is to apply during execution of the task. A set of possible constraints for the unconstrained pose parameter are selected for each unconstrained pose parameter. The sets of possible constraints are evaluated for the unconstrained pose parameters with respect to one or more task execution criteria. A nominal pose is determined for the tool based on a result of evaluating the sets of possible constraints for the unconstrained pose parameters with respect to the one or more task execution criteria. The robotic system is then directed to execute the task, including positioning the tool according to the nominal pose.

To control a motion of a device subject to constraints, a sequence of states and corresponding control inputs are transformed into a lifted space to determine a linear model of the dynamics of the device in the lifted space by minimizing fitting errors between the lifted states and approximation of the lifted states according to the linear control law. The fitting errors define an error model as a function bounding a data-driven envelope of a Lipschitz continuity on the fitting errors allowing to solve an optimal control problem in the lifted space according to the linear model subject to the constraints reformulated based on an evolution of the error model. The control input in the lifted space is transformed back to the original space for control.

Learning to effectively imitate human teleoperators, even in unseen, dynamic environments is a promising path to greater autonomy, enabling robots to steadily acquire complex skills from supervision. Various motion generation techniques are described herein that are rooted in contraction theory and sum-of-squares programming for learning a dynamical systems control policy in the form of a polynomial vector field from a given set of demonstrations. Notably, this vector field is provably optimal for the problem of minimizing imitation loss while providing certain continuous-time guarantees on the induced imitation behavior. Techniques herein generalize to new initial and goal poses of the robot and can adapt in real time to dynamic obstacles during execution, with convergence to teleoperator behavior within a well-defined safety tube.

A robot motion planning technique using an external computer communicating with a robot controller. A camera or sensor system provides input scene information including start and goal points and obstacle data to the computer. The computer plans a robot tool motion based on the start and goal points and the obstacle environment, where the robot motion is planned using either a serial or parallel combination of sampling-based and optimization-based planning algorithms. In the serial combination, the sampling method first finds a feasible path, and the optimization method then improves the path quality. In the parallel combination, both sampling and optimization methods are used, and a path is selected based on computation time, path quality and other factors. The computer converts dense planned waypoints to sparse command points for transfer to the robot controller, and the controller computes robot kinematics and interpolation points and controls the movement of the robot.

Systems, methods, devices, and other techniques are described for planning motions of one or more robots to perform at least one specified task. In some implementations, a task to execute with a robotic system using a tool is identified. A partially constrained pose is identified for the tool that is to apply during execution of the task. A set of possible constraints for the unconstrained pose parameter are selected for each unconstrained pose parameter. The sets of possible constraints are evaluated for the unconstrained pose parameters with respect to one or more task execution criteria. A nominal pose is determined for the tool based on a result of evaluating the sets of possible constraints for the unconstrained pose parameters with respect to the one or more task execution criteria. The robotic system is then directed to execute the task, including positioning the tool according to the nominal pose.

An object handling control device includes one or more processors configured to acquire at least object information and status information representing an initial position and a destination of an object; set, when a grasper grasping the object moves from the initial position to the destination, a first region, a second region, and a third region in accordance with the object information and the status information; and calculate a moving route along which the object is moved from the initial position to the destination with reference to the first region, the second region, and the third region.

A controller controls a motion of an object performing a task for changing a state of the object from a start state to an end state while avoiding collision of the object with an obstacle according to an optimal trajectory determined by solving an optimization problem of the dynamics of the object producing an optimal trajectory for performing the task subject to constraints on a solution of first-order stationary conditions modeling a minimum distance between the convex hull of the object and the convex hull of the obstacle using complementarity constraints.

A controller for controlling an operation of a robot to execute a task is provided. The controller comprises a memory configured to store a set of dynamic movement primitives (DMPs) associated with the task. The set of DMPs comprise a set of at least two dynamical systems: a function representing point attractor dynamics and a forcing function corresponding to a learned demonstration of the task. The controller comprises a processor configured to transform the set of DMPs to a set of constrained DMPs (CDMPs) by determining a perturbation function associated with the forcing function. The perturbation function is associated with a set of operational constraints. The processor is further configured to solve, a non-linear optimization problem for the set of CDMPs based on the set of operational constraints and generate, a control input for controlling the robot for executing the task, based on the solution.

Systems, methods, devices, and other techniques are described for planning motions of one or more robots to perform at least one specified task. In some implementations, a task to execute with a robotic system using a tool is identified. A partially constrained pose is identified for the tool that is to apply during execution of the task. A set of possible constraints for the unconstrained pose parameter are selected for each unconstrained pose parameter. The sets of possible constraints are evaluated for the unconstrained pose parameters with respect to one or more task execution criteria. A nominal pose is determined for the tool based on a result of evaluating the sets of possible constraints for the unconstrained pose parameters with respect to the one or more task execution criteria. The robotic system is then directed to execute the task, including positioning the tool according to the nominal pose.

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.