Ambi Robotics, Inc.

University of California

Matthew Matl

Stephen McKinley

David Gealy

Jeffrey Mahler

Kenneth Y Goldberg

Jeffrey Brian Mahler

Aaron Smith

A system and method for a dynamic robot kitting line that can include: processing a set of order requests and setting a packing fulfillment plan process for a robotic kitting line, wherein robotic kitting line comprises a conveyor system, a set of robotic workcells arranged along the conveyor system, and where each robotic workcell includes at least one robotic pick-and-place machine and a set of item bins; and managing operation of the robotic kitting line according to the packing fulfillment plan, which includes: conveying item totes through the set of robotic workcells, and for each item tote, progressively packing items of an order request assigned to an item tote by incrementally packing items at robotic workcells of the set of robotic workcells.

System and method for a dynamic robotic kitting line

A system and method for item sorting that can include an item selection unit with a robotic pick-and-place machine positioned adjacent to a defined input item region; an item sortation unit comprising a horizontal multi-degree-of-freedom gantry system and an item tote array, wherein the horizontal gantry system is positioned above the item tote array; wherein the item sortation unit further comprises an item holding and depositing system that is coupled to and actuated by the gantry system. The system and method can singulate an item from a collection of items and sort into one of a set of item totes.

System and method for robotic horizontal sortation

Embodiments are generally directed to generating a training dataset of labelled examples of sensor images and grasp configurations using a set of three-dimensional (3D) models of objects, one or more analytic mechanical representations of either or both of grasp forces and grasp torques, and statistical sampling to model uncertainty in either or both sensing and control. Embodiments can also include using the training dataset to train a function approximator that takes as input a sensor image and returns data that is used to select grasp configurations for a robot grasping or targeting mechanism.

Robotic systems and methods for robustly grasping and targeting objects

Growing community of inventors

El Cerrito, CA, United States of America

Matthew Matl