An object pose estimation system, an execution method thereof and a graphic user interface are provided. The execution method of the object pose estimation system includes the following steps. A feature extraction strategy of a pose estimation unit is determined by a feature extraction strategy neural network model according to a scene point cloud. According to the feature extraction strategy, a model feature is extracted from a 3D model of an object and a scene feature is extracted from the scene point cloud by the pose estimation unit. The model feature is compared with the scene feature by the pose estimation unit to obtain an estimated pose of the object.

A robot controller is a controller which controls, via a hand control device, a robot hand that grips an article with two or more gripping portions. The robot controller includes, a size information acquisition unit which acquires size information about the article based on an image obtained by a visual sensor for detecting the article, and a gripping adjustment unit which changes, in response to the size information, a gripping distance, which is the space between the gripping portions, in a gripping state or a gripping force of the gripping portions in the gripping state.

A model prediction control part of a control device includes an obstacle avoidance control unit that operates when there are a plurality of actual obstacles to be avoided. The obstacle avoidance control unit decides the position of a virtual obstacle from the positions of the plurality of actual obstacles acquired by an acquisition part so as to be positioned between the plurality of actual obstacles, and performs model prediction control by using, as the stage cost, the addition result of a standard cost and a virtual obstacle evaluation term for which a prescribed function, which uses, as parameters, at least the position of the virtual obstacle and the position of a moving body, is multiplied by a virtual obstacle weight. Using this configuration, when the moving body is caused to follow with respect to a target trajectory by the model prediction control, a collision with an obstacle can be avoided suitably.

Systems and methods of using a robot to train the system for motion detection are provided. A simple robot can be put in a room and programmed not to move except in accordance with specific programmed command. Such commands may be sent to the robot regarding movement(s) at a certain rate than could be seen in the response to a channel. A data set may be built over time, where the robot may be programmed to move such that the robot does change at specific times in duration and amount. Such robot motion may also be iterated. The algorithm records the impulse response changes associated with the robot changes and a database may be built based on such recorded and associated changes.

A machine learning system builds and uses computer models for controlling robotic performance of a task. Such computer models may be first trained using feedback on computer simulations of the robotic system performing the task, and then refined using feedback on real-world trials of the robot performing the task. Some examples of the computer models can be trained to automatically evaluate robotic task performance and provide the feedback. This feedback can be used by a machine learning system, for example an evolution strategies system or reinforcement learning system, to generate and refine the controller.

A method includes receiving, by a control system of a robotic device, data about an object in an environment from a remote computing device, where the data comprises at least location data and identifier data. The method further includes, based on the location data, causing at least one appendage of the robotic device to move through a predetermined learning motion path. The method additionally includes, while the at least one appendage moves through the predetermined learning motion path, causing one or more visual sensors to capture a plurality of images for potential association with the identifier data. The method further includes sending, to the remote computing device, the plurality of captured images to be displayed on a display interface of the remote computing device.

Disclosed herein is a robot including an output interface including at least one of a display or a speaker, and a processor configured to acquire output data of a predetermined playback time point of content output via the robot or an external device, recognize a first emotion corresponding to the acquired output data, and control the output interface to output an expression based on the recognized first emotion.

The present disclosure provides a conveyance system and the like capable of preferably conveying a conveyed object in accordance with a state of the conveyed object. The conveyance system includes a conveyance robot, a drive controller, which is a controller, an image data acquisition unit, and a setting unit. The conveyance robot conveys the conveyed object. The drive controller controls an operation of the conveyance robot. The image data acquisition unit acquires image data obtained by capturing images of the conveyed object. The setting unit sets an operation parameter of the conveyance robot in the drive controller based on the acquired image data.

A method for selecting an AI solution for an automated robotic process including receiving at least one functional media including information indicative of brain activity by a human engaged in a task of interest, analyzing the functional media, identifying an activity level in at least one brain region, identifying a brain region parameter and an activity parameter; identifying an action parameter based in part on the brain region parameter or the activity parameter; and selecting a component of the AI solution in part on the brain region parameter, the activity parameter, or the action parameter.

One variation of a method for autonomously scanning and processing a part includes: accessing a part model representing a part positioned in a work zone adjacent a robotic system; retrieving a sanding head translation speed; retrieving a toolpath for execution on the part defining positions, orientations, and target forces applied by the sanding head to the part. The method includes traversing the sanding head along the toolpath, at the sanding head translation speed; reading a sequence of applied forces from a force sensor coupled to the sanding head at positions along the toolpath; and deviating from the toolpath to maintain the set of applied forces within a threshold difference of a sequence of target forces along the toolpath. In one variation of the method, the robotic system executes a toolpath at a duration less than target duration by selectively varying target force and sanding head translation speed across the part.