Behavior Trees


The pre-print of our book titled Behavior Trees in Robotics and AI is available here.  The book is published by CRC Press Taylor & Francis. You can buy the hardcover copy and the ebook either on the CRC Press Store or Amazon.


Behavior Trees in Robmosys:

We are introducing Behavior Trees in the RobMoSys framework and we are working towards the development of formal verification tools to assess the correctness of defined behaviors in form of Behavior Trees.


Our open source library in ROS:

We have published our open source library for Behavior Trees in both github (link) and (link).

Blended Acting and Planning using Behavior Trees

We have proposed a framework that automatically synthesize a Behavior Tree that satisfy a given goal. it Is an attempt to  address the challenges regarding blending planning and acting: hierarchical organized deliberation and continual planning and deliberation.


Reinforcement Learning Using Behavior Trees:

Taking advantage of modularity and reactiveness of Behavior Trees,, we propose a model-free Automated Planned framework using Genetic Programming to derive an action plan for an autonomous agent to achieve a given goal in unknown environments. The advantages of the proposed approach is based on the advantages of BT over a genera FSM. In particular our approach avoids the problem of logic violation during the learning process.

Behavior Trees for Multi-Agent Systems:

By extending the single robot BT to a multi-robot BT, we are able to combine the fault tolerant properties of the BT, in terms of built-in fallbacks, with the fault tolerance inherent in multi-robot approaches, in terms of a faulty robot being replaced by another one. Furthermore, we improve performance by identifying and taking advantage of the opportunities of parallel task execution, that are present in the single robot BT. The extension of a single agent BT into a multi agent BT does not suffer of curse of dimensionality as a finite state machine does.


Stochastic Behavior Trees:

Stochastic Behavior Trees are an extension of Behavior Trees presented at ICRA 2014 where each leaf node is described by its success/failure probabilities and execution times. The recursive structure of the tree then enables us to step by step propagate such probabilities and time from the leaves to the root.

Maked and using for LenSlider e76736ab8e banner. Dont delete from media library.



Structural Properties of Behavior Trees:

Using standard tools of robot control theory, it is possible to evaluate structural properties such as robustness and safety of an action plan described by a Behavior Tree. In IROS 2014 we presented how to build action plans that are safe and robust.

Screen Shot 2015-03-18 at 10.38.37