Coordinating multiple robots to meet workflow and avoid conflict

A method, comprising:

• receiving sensor data via a communication interface, the sensor data including image data associated with a workspace;
• using the sensor data to generate a three dimensional view of at least a portion of the workspace, the three dimensional view including boundaries of a plurality of items present in the workspace;
• determining that no available grasp strategies exist for grasping an item among the plurality of items in the workspace;
• in response to a determination that no available grasp strategies exist for grasping an item in the workspace,
• determining a plan to operate robotic structure to change a state of the plurality of items or a flow of items into the workspace;
• causing the robotic structure to implement the plan to change a state of the plurality of items or a flow of items into the workspace;
• and
• in response to changing the state of the plurality of items or the flow of items, for each item of at least a subset of items, determining a grasp strategy for a current state of the item;
• for each grasp strategy, determining probability of grasp success, wherein the corresponding probability of grasp success is computed based at least in part on learned data corresponding to historical attempts to grasp a similar item;
• and
• using the grasp strategies and corresponding probabilities of grasp success to determine and implement a plan to autonomously operate a robotic structure to pick one or more items from the workspace and place each item singly in a corresponding location in a singulation conveyance structure.

A computer program product embodied in a non-transitory computer readable medium, comprising computer instructions for:

• receiving sensor data via a communication interface, the sensor data including image data associated with a workspace;
• using the sensor data to generate a three dimensional view of at least a portion of the workspace, the three dimensional view including boundaries of a plurality of items present in the workspace;
• determining that no available grasp strategies exist for grasping an item among the plurality of items in the workspace;
• in response to a determination that no available grasp strategies exist for grasping an item in the workspace,
• determining a plan to operate robotic structure to change a state of the plurality of items or a flow of items into the workspace;
• causing the robotic structure to implement the plan to change a state of the plurality of items or a flow of items into the workspace;
• and
• in response to changing the state of the plurality of items or the flow of items, for each item of at least a subset of items, determining a grasp strategy for a current state of the item;
• for each grasp strategy, determining a probability of grasp success, wherein the corresponding probability of grasp success is computed based at least in part on learned data corresponding to historical attempts to grasp a similar item;
• and
• using the grasp strategies and corresponding probabilities of grasp success to determine and implement a plan to autonomously operate a robotic structure to pick one or more items from the workspace and place each item singly in a corresponding location in a singulation conveyance structure.