Dexterity, Inc.

Shitij Kumar

Andrew Lovett

Kevin Jose Chavez

Rohit Arka Pidaparthi

Cyril Nader

William Arthur Clary

Jeesu Baek

Jonathan Kuck

Neeraja Abhyankar

Zhouwen Sun

Robert Holmberg

Joseph Li

Kishore Ranganath Ramakrishnan

Ben Varkey Benjamin Pottayil

Alberto Leyva Arvayo

David Leo Tondreau, Iii

Farshid Farhat

Varun Uday Nayak

Yuan Gao

Salvador Perez

Robert Moreno

Justin Sunjoon Kim

Arth Beladiya

A robotic system is disclosed. The system includes a communication interface configured to receive, from one or more sensors deployed in a workspace, sensor data indicative of a current state of the workspace, the workspace comprising a pallet or other receptacle and a plurality of items stacked on or in the receptacle. The system includes one or more processors that use a geometric model based at least in part on past item placements in combination with the sensor data to estimate a state of the pallet or other receptacle and one or more items stacked on or in the pallet or other receptacle, and use the estimated state to generate or update a plan to control a robotic arm to place a next item on or in, or remove a next item from, the pallet or other receptacle in a manner that avoids having the next item collide with any other item stacked on or in the pallet or other receptacle.

Collision avoidance based on vision data, geometric data and physics engine evaluation

A robotic system is disclosed. The system includes a memory that stores a machine learning-based model to provide a scoring function value for a candidate item placement on a pallet on which are plurality of items are to be stacked given a current state value of the pallet and a set of zero or more items placed previously. The system includes one or more processors that use the model to determine a corresponding score for each of a plurality of candidate placements for a next item to be placed and the current state value associated with the current state of the pallet and a set of zero or more items placed previously, select a selected placement based at least in part on the respective scores, control a robotic arm to place the next item according to the selected placement.

Workflow for using learning based approach for placing boxes on pallets

An end effector is disclosed. The end effector includes a first set of one or more suction cups of a first size having a first diameter, and a second set of one or more suction cups of a second size having a second diameter that is smaller than the first diameter. The end effector further includes a first actuation mechanism configured to apply suction to at least a subset of the first set of suction cups independently of actuation of the second set of suction cups and a second actuation mechanism configured to apply suction to at least a subset of the second set of suction cups independently of actuation of the first set of suction cups.

Suction-based end effector with mixed cup sizes

A robotic system is disclosed. The system includes a communication interface that receives, from one or more sensors deployed in a workspace, sensor data indicative of a current state of the workspace. The system includes one or more processors that use the sensor data to estimate a state of one or both of the pallet or other receptacle and the set of zero or more items stacked on or in the receptacle, and use the estimated state to generate or update a plan to control a robotic arm to place a next set of items on or in, or remove the next set of items from, the pallet or other receptacle, the plan comprising an ordered sequence of item placements or removals. The plan is generated or updated based at least in part by performing a bounded tree search in which a subset of possible ordered sequences is explored.

Workflow for using tree search-based approach for placing boxes on pallet with limited knowledge of future sequence

A method and system for obtaining an identifier from an item is disclosed. The method includes autonomously operating a robotic structure having a robotic arm to pick an item using an end effector of the robotic arm along a predetermined path from a source location to a destination location according to a plan. The picked item is moved according to the plan. An active measure is determined to be performed at least in part to obtain an identifier determined to be missing based on information received via the communication interface from one or more sensors. The robotic structure is autonomously operated to place the item at the destination location based at least in part on the plan.

Robotic system for identifying items

A task table for use in a robotic system comprising a robot mounted on a carriage configured to move the robot along a path is disclosed. The task table includes a substantially horizontal table surface, and one or more mounting arms to which the table surface is affixed at a first distal end and comprising, at a second distal end opposite the first distal end, one or more mounting structures, the one or more mounting arms having a length that allows the table surface to be mounted to the robot or the carriage at a distance from a base of the robot that enables the robot to reach a plurality of locations on the table surface each with a desired pose.

Robot carriage tray table

A robotic system is disclosed. The system includes a memory that stores for each of a plurality of items a set of attribute values. The system includes a processor(s) that uses the attribute values to simulate the placement of items, including by determining, iteratively, for each next item a placement location at which to place the item on a simulated stack of items on the pallet, using the attribute values and a geometric model of where items have been simulated to have been placed to estimate a state of the stack after each of a subset of simulated placements, and using the estimated state to inform a next placement decision. The steps of determining for each next item a placement location and estimating the state of the stack until all of at least a subset of the plurality of items have been simulated as having been placed on the stack.

Simulated box placement for algorithm evaluation and refinement

A robotic system is disclosed. The system includes a communication interface that receives, from a sensor(s) deployed in a workspace, sensor data indicative of a current state of the workspace, the workspace comprising a pallet or other receptacle and a plurality of items stacked on or in the receptacle. The system includes one or more processors that control a robotic arm to place a first set of items on or in, or remove the first set of items from, the pallet or other receptacle, update a geometric model based on the first set of items placed on or in a receptacle, use the geometric model in combination with the sensor data to estimate a stack of one or more items on or in the receptacle, and use the estimated state to generate or update a plan to control the robotic arm to place a second set of items.

State estimation using geometric data and vision system for palletizing

A robotic end effector is disclosed. The end effector is configured to grasp, move, and place one or more objects without assistance from another robot. The end effector includes a lateral member configured to be attached to a robotic arm at a central portion of the lateral member, a passive side member fixedly mounted to the lateral member at a first distal end and configured to engage mechanically with a first recess on a first side of an object to be grasped, and an active side member hinged to the lateral member at a second distal end opposite the first distal end and configured to engage mechanically with a second recess on a second side of the object to be grasped, the second side being opposite the first side of the object to be grasped. The passive side member and the active side member each include a structure, and the structure on the passive side member and the active side member have different profiles.

Robotic tray gripper

A method and system for obtaining an identifier from an item is disclosed. The method includes autonomously operate a robotic structure to move an item along a predetermined path from a source location to a destination location, and autonomously operating the robotic structure to place the item at the destination location based at least in part on the plan. The item comprises one or more identifiers, and in response to a determination that at least one of the one or more identifiers was not obtained by one or more sensors, an active measure is performed to cause the one or more sensors to obtain the at least one identifier that was not obtained. The predetermined path corresponds to a path along which the item is moved from the source location to the destination location. The predetermined path is planned so that the item is moved within a threshold range of the one or more sensors while the item is moved along the predetermined path.

A robotic end effector is disclosed. The end effector is configured to grasp, move, and place one or more objects without assistance from another robot. The end effector includes a lateral member configured to be coupled to a robotic arm, a passive side member coupled substantially rigidly to the lateral member at a first distal end and configured to engage mechanically with a first recess on a first side of an object to be grasped, and an active side member coupled to the lateral member at a second distal end opposite the first distal end and configured to engage mechanically with a second recess on a second side of the object to be grasped, the second side being opposite the first side of the object to be grasped. The passive side member and the active side member each include a structure, and the structure on the passive side member and the active side member have different profiles.

Growing community of inventors

Redwood City, CA, United States of America

Shitij Kumar