technical advisor : mr. roni stern academic advisor : dr. meir kalech team members : amit ofer ...
Post on 19-Dec-2015
217 views
TRANSCRIPT
The MPAS Project Multi-agent Pathfinding Algorithms Simulator
Technical Advisor : Mr. Roni Stern Academic Advisor : Dr. Meir KalechTeam members :
Amit Ofer Liron Katav
Project Homepage : http://mpasproject.weebly.com
Introduction
Path finding refers to the problem of searching the shortest route between two points.
Multi-agent path finding problem involves navigating units from their starting position to their respective goals, whilst going around any static obstacles and other moving units along the way.
Introduction
The problem is becoming increasingly
important in many real-life applications, including motion planning in robotics, air traffic control, vehicle routing, military operation planning and computers games.
Problem domain
The standard algorithm for this problem is the A-star (A*) algorithm.
A star is an extension of Dijkstra’s algorithm, A* achieves better performance (with respect to time) by using heuristics.
Problem domain
The problem with the A-star algorithm is that its complexity grows exponentially with the number of mobile units on the map, making it not practical for real time applications.
For this reason the modern research focuses on finding a more efficient algorithms that solve the multi-agent pathfinding problem.
Vision
Our goal is to develop a simulator that will help to observe the different behaviors and compare the performance of various multi-agent pathfinding algorithms.
The algorithms that will be tested are: A-star (A*), 1968. Hierarchical Cooperative A* (HCA*) ,2005. Operator Decomposition + Independence
Detection, 2010.
Vision
The algorithms will be tested on two environments:
Grid map – a tiled based map where each unit can move to one of the 8 adjacent tiles.
Geographical map – a real world map where the mobile units are limited to moving on the roads.
System Architecture - High level
Geographical map
environment
Grid map environment
External GIS
MPAS
System architecture
Algorithm Layer
Presentation layer
Controller Layer
-Method Invocation-Events
User input
ViewChanges
Input
State Change
Main Functional Requirements
Choose the number of agentsIn the grid map environment:
Choose the size of the grid mapIn the geographical map environment:
Choose the mapLoad grid mapsSave grid mapsClear mapSets the starting and finishing cells for each agentSet blocking cells
Main Functional Requirements
Choose the algorithm to be testedChoose the heuristic to be usedStart the simulationStop the simulationRunning the simulation Step by stepGenerate random scenarioRestart simulation
Main Non Functional Requirements
Speed The system should launch in less than 1 minute. It gives an output in no more than 15 minutes (for an average
problem’s size). Capacity
Up to 1 Million vertices (1000 *1000 on grid or 1 Million on geographical-map)
Up to 100 agents that will run simultaneously. Portability
The system should operate on Linux and Windows (XP/Vista/7). The system should be able to run on a standard pc computer
(though calculation times may vary according to system specs).
Main Non Functional Requirements Usability
The system GUI should be user-friendly and easy to use. The system should be simple to manage for the common
user. The learning pace of the system should be quick.
Availability The system should be able to operate at any time of day
and no matter the amount of applications running at the background of the Operating System.
Main Non Functional Requirements Extensibility
All algorithms will implement a predefined interface. Thus the simulator will be easy to extend by adding more algorithms that will implement this interface.
Platform Constraints The application will be developed in Java. The computer that will run the system should not be older
than 3 years and include JRE and java version 1.6 or higher.
Project’s status
ARD
Prototype v1.0 includes :
• A-star implementation with 2 agents on a grid environment
References
A- Star alorithm http://en.wikipedia.org/wiki/A_star
D. Silver, 2005. Cooperative Pathfinding. T. Standley, 2010. Finding Optimal
Solutions toCooperative Pathfinding Problems .