The present disclosure is for systems and methods for adjusting operational configurations of robots in real-time. The invention pertains to overriding or replacing one operational configuration of a robot with another when appropriate circumstances arise and certain conditions have been met. In one aspect, the invention is applicable to robotic picking operations and serves to allow for unique robotic picking operations outside of the normal or standard limitations typically imposed on a robotic picking system. The invention provides the ability to remotely adjust robotic operational configurations in real-time, on-demand, in order to address various circumstances that may arise without requiring interruption of a picking session or requiring on-site human intervention.

The technology is directed to training a system to generate pick instructions. A teleoperator system may receive data corresponding to a robot attempting a picking task including picking an item of an identified product type from a container. The data may include imagery of an end effector of the robot after the attempted picking task. The teleoperator system may display the imagery on a display and an input indicating whether the picking task was successfully or unsuccessfully performed by the robot may be received. The data may be labeled based on the input and transmitted to a processor for training a learning algorithm for use in generating future pick instructions.

The present invention relates to systems and methods for accounting for edge cases (i.e. tail data) in automated decision making systems, for example automated robotic picking systems. The systems and methods provide for retraining machine learning (ML) models so that the edge cases can be handled in a manner that requires less (or no) human intervention. The disclosed systems and methods create updated ML models, replacement ML models, and/or supplementary ML models that can provide better performance (e.g. improved automated robotic picking) when edge cases are encountered. Furthermore, the present inventions disclose systems and methods for obtaining training data faster and in a more cost effective manner, which enables the systems and methods disclosed herein to update models at a faster rate, thereby enabling broader, system-wide handling of edge cases in a more effective and efficient manner.

A picking apparatus in an embodiment includes: a gripper, an arm, a detector, and a control unit. The gripper picks and grips an object to be conveyed. The arm moves the gripper and causes the gripper to convey the object to be conveyed. The detector is attached to the arm and senses a force applied to the gripper. The control unit controls an operation of the gripper and the arm. The control unit includes a calculator and a subtractor. The calculator calculates a gravitational force and an inertial force applied to the gripper when the gripper grips and moves the object to be conveyed using an arithmetic expression including a coefficient determined in accordance with a mass of the object to be conveyed. The subtractor subtracts the gravitational force and the inertial force calculated by the calculator from a force applied to the gripper sensed by the detector.

Autonomous operations, such as robotic grasping and manipulation, in unknown or dynamic environments present various technical challenges. For example, three-dimensional (3D) reconstruction of a given object often focuses on the geometry of the object without considering how the 3D model of the object is used in solving or performing a robot operation task. As described herein, in accordance with various embodiments, models are generated of objects and/or physical environments based on tasks that autonomous machines perform with the objects or within the physical environments. Thus, in some cases, a given object or environment may be modeled differently depending on the task that is performed using the model. Further, portions of an object or environment may be modeled with varying resolutions depending on the task associated with the model.

A robot and a robot-based container storage and removal method. The robot comprises: a master control processing unit (110), a pick-and-place mechanism (120) and a marker detection unit (130), wherein according to target storage and removal position information of a target inventory container, the master control processing unit (110) controls a robot body to move to a first horizontal position and controls the pick-and-place mechanism (120) to move to a first height position; when the robot body and the pick-and-place mechanism (120) stop moving, the marker detection unit (130) determines a target pick-and-place marker from a target inventory support to which the target inventory container belongs; and the master control processing unit (110) also calibrates the position of the pick-and-place mechanism (120) according to the position of the target pick-and-place marker, so as to control the calibrated pick-and-place mechanism (120) to perform a storage operation or a removal operation on the target inventory container. By means of the solution, a pick-and-place position of a pick-and-place mechanism (120) of the robot can be precisely positioned and moved, such that the pick-and-place mechanism (120) can quickly and accurately store or remove a target inventory container.

A machine learning method for learning an action of a robot including a hand to pick out a workpiece from a container containing a plurality of the workpieces stacked in bulk and install the workpiece such that the workpiece is in a predetermined installation state includes learning a reverse-order action of removing, by the hand, the workpiece in the predetermined installation state after completion of installation, and learning an installation order of the workpiece based on a learning result of the reverse-order action of removing the workpiece.

A photographing method for picking or placing, applied includes: obtaining first multi-dimensional image information of a target position in a target shelf; determining, according to the first multi-dimensional image information, whether there is a first item in the target position; and determining a photographing strategy of the photographing module according to a determining result, wherein the photographing strategy includes one of: the photographing module not moving with the handling apparatus in a telescopic direction for continued photographing, the photographing module moving a preset distance along the telescopic direction with the handling apparatus, and performing an operation of starting a solution; the solution includes at least one of stopping photographing, sending a warning signal, and reporting to a server to which the transport robot belongs.

An embodiment of the disclosure provides a warehouse robot control method and apparatus, a robot, and a warehouse system. The method includes: obtaining a container scheduling instruction of a first target container, where the container scheduling instruction includes a container type of the first target container; determining a container pose recognition algorithm according to the container type of the first target container, and recognizing pose information of the first target container based on the container pose recognition algorithm; and controlling the warehouse robot to pick up the first target container according to the pose information. For different types of containers, using corresponding pose recognition algorithms to determine pose information thereof and automatically picking up different types of the containers based on the pose information improve application range and efficiency of automatic pickup of the warehouse robot.

A method of processing O-rings in an automated mass production system includes: (a) advancing an O-ring retainer toward a loading position in alignment with an output end of a feed device; (b) discharging a leading O-ring from the output end in electronic synchronization with advancement of the O-ring retainer to the loading position to initiate loading of the O-ring into the retainer prior to the retainer arriving at the loading position; (c) after loading the O-ring into the retainer, advancing the retainer away from the loading position toward an unloading position; and (d) moving an end effector in electronic synchronization with advancement of the retainer to the unloading position to synchronize arrival of the retainer at the unloading position with arrival of the end effector at a pick-up position in alignment with the O-ring at the unloading position for pick up of the O-ring by the end effector.