Techniques described herein relate to using reduced-dimensionality embeddings generated from robot sensor data to identify predetermined semantic labels that guide robot interaction with objects. In various implementations, obtaining, from one or more sensors of a robot, sensor data that includes data indicative of an object observed in an environment in which the robot operates. The sensor data may be processed utilizing a first trained machine learning model to generate a first embedded feature vector that maps the data indicative of the object to an embedding space. Nearest neighbor(s) of the first embedded feature vector may be identified in the embedding space. Semantic label(s) may be identified based on the nearest neighbor(s). A given grasp option may be selected from enumerated grasp options previously associated with the semantic label(s). The robot may be operated to interact with the object based on the pose and using the given grasp option.

Embodiments generally relate to robots and enabling robots to locate objects in a physical environment. In some embodiments, a method includes charging a radio-frequency identification (RFID) tag with an RFID reader, where the RFID tag is coupled to an object, and where the RFID reader is coupled to a robot arm. The method further includes receiving a plurality of responses from the RFID tag, where each response includes a power value to which the RFID tag was charged and a time value for charging the RFID tag to the power value. The method further includes moving the RFID reader to a plurality of RFID reader positions using the robot arm, where each RFID reader position is associated with one of the responses of the plurality of responses. The method further includes determining a plurality of distances from the RFID reader to the RFID tag based on power values and the time values of the respective responses at the respective RFID reader positions. The method further includes determining a location of the RFID tag based on the plurality of distances.

A service rendering vehicle renders a service on a service object. A controller of the service rendering vehicle receives an image that includes a service label corresponding to the service object, captured by a camera coupled to the controller. The controller identifies a label image of the service label in the image. The controller obtains a service point location and a service request encoded in the service label. The service point location includes a location relative to a location of the service label. The controller calculates a label position of the service label using the image of the service label and calculates a service point position using the service point location and the label position. The controller obtains a service from the service request and issues instructions to move the service rendering vehicle to the service object and to render the service at the service point position.

A method includes: scanning a first QR code on a first physical object to obtain first parameters, wherein the first parameters include specifications for making a product using a manufacturing process, and the specifications include at least one material used to build the product. The method further includes scanning a second QR code on a second physical object to obtain second parameters, wherein the second parameters correspond to the manufacturing process; comparing, by a computing device, the first parameters and the second parameters; based on comparing the first parameters and the second parameters, determining whether the first object corresponds to the second object; and in response to determining that the first object corresponds to the second object, building the product using the at least one material.

A processing system is disclosed for processing objects. The processing system includes a perception system for providing perception data regarding an object, and a primary transport system for providing transport of the object along a primary direction toward a processing location that is identified based on the perception data.

Techniques described herein relate to using reduced-dimensionality embeddings generated from robot sensor data to identify predetermined semantic labels that guide robot interaction with objects. In various implementations, sensor data obtained from one or more sensors of a robot includes data indicative of an object observed in an environment in which the robot operates. The sensor data is processed utilizing a first trained machine learning model to generate a first embedded feature vector that maps the data indicative of the object to an embedding space. Nearest neighbor(s) of the first embedded feature vector is identified in the embedding space. Semantic label(s) are identified based on the nearest neighbor(s). A given grasp option is selected from enumerated grasp options previously associated with the semantic label(s). The robot is operated to interact with the object based on the pose and using the given grasp option.

The installation position pointer system comprises a laser source to supply a laser beam, a driving device to align the laser beam, a controller device to control the operation of the driving device in accordance with an installation position data (IPD) of an attachment to be installed in a building and an associated reference position in the building, and an input device to obtain the IPD of a plurality of attachments from the attachment. The IPD of the attachments may be provided by an attachment installation position database system.

A drop perception system is disclosed that includes an open housing structure having an internal volume, an open top and an open bottom, and a plurality of perception units positioned to capture perception data within the internal volume at a plurality of locations between the open top and the open bottom of the open housing.

A drop perception system is disclosed that includes an open housing structure having an internal volume, an open top and an open bottom, and a plurality of perception units positioned to capture perception data within the internal volume at a plurality of locations between the open top and the open bottom of the open housing.

A robotic system is disclosed that include an articulated arm and a first perception system for inspecting an object, as well as a plurality of additional perception systems, each of which is arranged to be directed toward a common area in which an object may be positioned by the robotic arm such that a plurality of views within the common area may be obtained by the plurality of additional perception systems.