Controlling process of robots having a behavior tree architecture
AssigneeMagazino GmbH
InventorMoritz Tenorth
Life of this patent
Estimating the term…
Abstract
The present invention relates to a method for controlling a robot, the method including usage of a behavior tree architecture for tasks performed by the robot. The present invention also relates to a system comprising a data processing means adapted to carry out the method, wherein the system preferably comprises a robot. The present invention also relates to a use of a behavior tree architecture for programming, supervision, introspection and/or debugging.
Claims14
A method for controlling a robot, the method including using a behavior tree architecture for tasks performed by the robot,
- wherein the robot includes a mobile base and/or a manipulator,
- wherein the behavior tree architecture comprises nodes including a root node, at least one control flow node and at least one execution node and directed edges,
- wherein the nodes are connected with the directed edges,
- wherein the root node has one outgoing edge and no incoming edge, each control flow node has one incoming edge and at least one outgoing edge and each execution node has one incoming edge and no outgoing edge, wherein a node having an outgoing edge is defined as a parent node in relation to another node connected to this edge and a node having an incoming edge is defined as a child node in relation to another node connected to this edge,
- wherein the control flow nodes and the execution nodes are adapted to return different states to their parents nodes, the states including success, running and failure,
- the method including the steps of
- the root node calling a control flow node;
- the control flow node calling a first node;
- the first node returning a first state to the control flow node;
- and
- the control flow node returning a second state to the root node;
- wherein the at least one control flow node comprises a composite task with multiple child tasks, wherein the composite task defines which data elements serve as its input and output, and
- wherein, in addition to returning the states to its parent node, the composite task writes values of its output data elements to its parent's data storage.
A method including controlling a robot and using a behavior tree architecture for tasks performed by the robot,
- wherein the robot is an autonomously driving robot,
- wherein the behavior tree architecture comprises nodes including a root node, at least one control flow node and at least one execution node and directed edges,
- wherein the nodes are connected with the directed edges,
- wherein the root node has one outgoing edge and no incoming edge, each control flow node has one incoming edge and at least one outgoing edge and each execution node has one incoming edge and no outgoing edge, wherein a node having an outgoing edge is defined as a parent node in relation to another node connected to this edge and a node having an incoming edge is defined as a child node in relation to another node connected to this edge,
- wherein the control flow nodes and the execution nodes are adapted to return different states to their parents nodes, the states including success, running and failure,
- the method including the steps of
- the root node calling a control flow node;
- the control flow node calling a first node;
- the first node returning a first state to the control flow node;
- and
- the control flow node returning a second state to the root node;
- wherein the at least one control flow node comprises a composite task with multiple child tasks, wherein the composite task defines which data elements serve as its input and output, and
- wherein, in addition to returning the states to its parent node, the composite task writes values of its output data elements to its parent's data storage.
A method including controlling a robot and using a behavior tree architecture for tasks performed by the robot,
- wherein the robot includes a sensor and the method includes sensing information by means of the sensor and effecting the robot to respond to the information,
- wherein the behavior tree architecture comprises nodes including a root node, at least one control flow node and at least one execution node and directed edges,
- wherein the nodes are connected with the directed edges,
- wherein the root node has one outgoing edge and no incoming edge, each control flow node has one incoming edge and at least one outgoing edge and each execution node has one incoming edge and no outgoing edge, wherein a node having an outgoing edge is defined as a parent node in relation to another node connected to this edge and a node having an incoming edge is defined as a child node in relation to another node connected to this edge,
- wherein the control flow nodes and the execution nodes are adapted to return different states to their parents nodes, the states including success, running and failure,
- the method including the steps of
- the root node calling a control flow node;
- the control flow node calling a first node;
- the first node returning a first state to the control flow node;
- and
- the control flow node returning a second state to the root node;
- wherein the at least one control flow node comprises a composite task with multiple child tasks, wherein the composite task defines which data elements serve as its input and output, and
- wherein, in addition to returning the states to its parent node the composite task writes values of its output data elements to its parent's data storage.
Dependencies are inferred from the claim text.