https://code.ros.org/svn/wg-ros-pkg/branches/trunk_cturtle/stacks/executive_trex/

The Teleo-Reactive Executive (T-REX) is a hybrid executive combining goal-driven and event-driven behavior in a unified framework based on temporal plans and temporal planning. T-REX uses a single declarative programming language and shared runtime state for both deliberation and execution offering a seamless integration between planning and control. T-REX also provides explicit support for compositional control which improves modularity, scalability, and robustness.

History

TREX was developed at the Monterey Bay Aquarium Research Institute for Adaptive Mission Control of Autonomous Underwater Vehicles. TREX has been deployed for a number of sea trials and science missions in the Monterey Bay. Some of the details of these field experiments are described here

TREX exploits the EUROPA-2 temporal planning library which is an open-source system from the NASA Ames Research Center applied to diverse problems including rover command and control, solar array control for the space station, crew activity planning and scheduling for the space station, and motion planning for the ATHLETE robot.

Documentation

Programming a robot to operate autonomously, competently and robustly in complex, dynamic and uncertain environments is really hard. This project aims to make it easier through a new programming paradigm based on high-level models and automated planning.

Our approach falls into the category of model-based programming. The basic idea is to build models of how a robot should behave, and then apply those models directly to achieve that behavior at run-time. This enables a very high-level approach to building goal-directed, and reactive agents, and provides a specification that a planner can naturally manipulate.

To implement this, we have built a run-time context capable of executing such model-based programs. That runtime is called an executive. We call T-REX a hybrid executive because it can support discrete and continuous states, which is important to do things in the real world with robots. It can also span a continuum between reactive and deliberative behavior. This is important because sometimes a robot must act quickly with limited consideration for future impacts, and sometimes it can deliberate over a range of possible actions and their implications in the future.

More information is available here that teaches the background concepts of TREX and helps the user understand more about the technology and how it works. This is a good place to go if you want to better understand how TREX might help you, and assess how it fits in with other technologies.

If you have any questions or comments about T-REX, mail trex-users@googlegroups.com, or visit the trex-users google group.

Join the Community

TREX is an open-source project that is open to new-comers and new ideas. We welcome new users and new contributors. This material offers pointers for getting involved and learning the ropes from a developers perspective.

*Planning version 1.0 for Q4*. After a summer of development, we are basically feature complete with a good regression test suite in place. We are now turning to documentation in preparation for a 1.0 release in Q4 of 2009.

Report a Bug

<<TracLink(REPO COMPONENT)>>