P-MAS-TG

Planner for Multiple Agent System with Temporal Goals

---

Description

This package contains implementation for plan synthesis algorithms given a finite transition system (as the agent motion model) and a Linear temporal logic formula (as the agent task). It outputs the static plan as a sequence of agent motion and action, required to fulfill the task.

---

Features

• Allow both normal and co-safe LTL task formulas.
• Action model can be muted if only motion is concerned.
• Soft specification is optional.
• NetworkX structure for FTS, Buchi and Product automata.
• Static or on-the-fly construction of the product automaton.
• Stand-alone planner.

from P_MAS_TG.ts import MotionFts, ActionModel, MotActModel
from P_MAS_TG.planner import ltl_planner

# construct your motion and action model
#---------
robot_motion = MotionFts(node_dict, symbols, 'your_ws_name')
robot_motion.set_initial(initial_node)
robot_motion.add_un_edges(edge_list, unit_cost = 0.1)
#---------
robot_action = ActionModel(action_dict)
#---------
robot_model = MotActModel(robot_motion, robot_action)

# specify your hard and soft tasks
hard_task = '(([]<> r3) && ([]<> r4))'
soft_task = None

# set planner
robot_planner = ltl_planner(robot_model, hard_task, soft_task)

# synthesis
robot_planner.optimal(10,'static')

• Generate .dat for MatLAB to load Buchi and product automata model. See square_world.py.

---

Debugging

• Install python packages like networkx, ply.
• Add this package to your PYTHONPATH, to import it in your own project.
• ltlba_32 and ltlba_64 are executable files complied under OS X. For other OS, please follow ltl2ba/README.txt.
• Try test.py and other examples in the Examples folder.