Swarming Micro Air Vehicle Networks SMAVNET
#1

PRESENTED BY:
T. M. Praneeth
G. Naresh

[attachment=9590]
Swarming Micro Air Vehicle Networks SMAVNET
Abstract

The SMAVNET project aims at developing swarms of flying robots that can be deployed in disaster areas to rapidly create communication networks for rescuers. Flying robots are interesting for such applications because they are fast, can easily overcome difficult terrain, and benefit from line-of-sight communication. This is inspired from biology to implement controllers based on ant-foraging or resulting from artificial evolution. Finally, the first steps towards the deployment
of aerial ad-hoc networks in reality are shown.
I. INTRODUCTION
Swarms of flying robots can be used in disaster areas to autonomously create communication networks for rescuers and victims (Fig. 1). Flying robots have the advantage of rapidly overcoming difficult terrain and providing unobstructed wireless communication. To allow for a swarm composed of cheap, transportable and robust robots, we avoid using positioning sensors which typically depend on the environment (GPS, cameras) or are expensive and heavy (lasers, radars). Instead, robot behaviors react to local wireless communication with robots within transmission range. Using the radio module itself for controlling the behavior of the robot (communication-based behavior) is appealing since it directly relates to the capacity of the robot to send and receive radio messages.
However, there currently exists no methodology to design robot controllers resulting in the emergence of desired swarm behaviors. Here, we propose two bio-inspired techniques to overcome this problem. In the first approach, we use artificial evolution as a mean to automatically design simple, efficient and unthought-of controllers for robots. We then reverse engineer these controllers and reuse the discovered principles in different scenarios. In the second approach, we look at the creation, maintenance and evaporation of army-ant pheromone trails during foraging and apply the same principles to the design of robot controllers for the deployment, maintenance and retraction of communication networks.
Finally, a first step towards experiments in reality by showing the steering of a single flying robot using only communication hardware (e.g. WiFi module or radio modem) and the current fleet of robots being developed in the scope of the SMAVNET project.
Fig. 1 Artistic view of the use of a group of flying robots for establishing communication networks between rescuers on the ground in a flood scenario.
Laboratory of Intelligent Systems, Ecole Polytechnique
F´ed´erale de Lausanne, Switzerland {sabine.hauert,
II. EVOLVED NETWORK DEPLOYMENTS
Artificial evolution has been extensively used for the development of robot controllers due to its capacity to automatically engineer solutions displaying complex abilities using simple and efficient behaviors. Systems of interest generally can not be solved using conventional programming techniques because they are highly non-linear, stochastic or poorly understood. Subsequently, artificial evolution is particularly well suited for the design of controllers for swarms of robots. In addition, the mechanisms leading to the evolution of cooperation in natural and robotic systems have just recently been understood. In particular, genetic algorithms and genetic programming have successfully been used to design controllers for swarms of ground and aerial vehicles in simulation or on-board physical robots in research environments.
Designing swarm controllers for flying robots is especially challenging because of the lack of positioning information, which is unprecedented in the literature, and the dynamics of our fixed-wing flying robots that must always remain in motion to avoid stalling. To overcome these challenges, we use artificial evolution as a means to automatically design neural controllers for the robots.
An example showing the behavior of the evolved swarm forming an ad-hoc network between two rescuers can be seen in Fig. 2. The strategy adopted by the swarm consists in forming a tight chain of robots which grows as long as additional robots are launched from the rescuer to the South.
Fig. 2. Trajectories performed by the evolved robots during a 30 min mission. In this mission, robots are launched from a rescuer at regular intervals and must self-organize to search for the second rescuer to the North-East. Robots form chains that can translate from West to East until the second rescuer is found. Robots then maintain the connection by turning on the spot. The trajectory of the first launched robot is shown by a light grey line.
Once all flying robots have been launched, the chain shifts along the communication range of the launching rescuer, sweeping the area from West to East until a second rescuer is found. The communication page link between the two rescuers is then maintained by having all robots turn on the spot with the smallest possible radius given the dynamics of the aircraft.
However, evolved controllers are often unable to adapt across different scenarios without being re-evolved. This process takes time and is unrealistic for robot swarms which are intended to be used out-of-the-box in critical applications. Instead we propose to reverse-engineer evolved controllers so as to capture the simplicity and efficiency found through evolution in hand-designed robot controllers whose parameters can easily be optimized for various scenarios. For our application, reverse-engineered controllers resulted in three simple local-interactions responsible for the emergent behavior of the swarm, namely chain formation, translation and communication maintenance. Such rules form the basis for controllers which will be adapted to real-life scenarios with wind, varying robot dynamics or mobile rescuers.
III. ANT-BASED NETWORK DEPLOYMENTS
Army ant colonies display complex foraging raid patterns involving thousands of individuals communicating through chemical trails (pheromone). These structures are thought to reflect an optimized mechanism to explore and exploit food resources in nature. By taking inspiration from the foraging mechanism found in army ants, we want to create, maintain and retract aerial ad-hoc networks between rescuers. However, in real-life applications, it is often undesirable to modify the environment in which robots deploy (by physically depositing chemicals or objects) and the deploying substrate is often unstable (e.g., air, water and quickly modifiable environments). Also, depositing virtual pheromone on a map is not possible when no global positioning is available. To solve this issue in our system, pheromone is virtually deposited on the robots (pheromone robotics). The approach proposed here consists of separating the flying robots into two types, namely “nodes” and “ants”. Nodes constitute the environment on which pheromone can be virtually deposited and read from. Ants are capable of navigating through a grid of nodes while depositing virtual pheromone on them through the use of local wireless communication. Furthermore, robots can dynamically change between both categories.
Fig. 3. Simulator screenshot showing the successful ant-based deployment of flying robots forming an ad-hoc network between two rescuers. Nodes are white with black borders, ants are in solid black and lines represent local communication links.
An example of an ant-based swarm behavior in simulation can be seen in Fig. 3. Observed behaviors include the formation of grids composed of several short branches deployed in multiple directions or longer chain-like grids capable of searching in a single direction for distant rescuers.
The overall network changes between different configurations until a rescuer is found. The network is then optimized and maintained by attracting robots to useful positions in the network using pheromone. Finally, because pheromone evaporates, robots eventually retract to the nest where they are either told to redeploy or land.
IV. EXPERIMENTAL SETUP
The communication-based behavior was implemented on board a fixed-wing aerial robot developed in the scope of the SMAVNET project1 and now commercialized by senseFly2. Tests were conducted outdoors in a fully autonomous manner. We hereby introduce the flying platform, the base station and the radio module used in the experiments presented in this paper.
A. Flying Platform
We use a light weight (420 g) and safe flying platform that has been developed for the rapid-prototyping of aerial robot experiments. The flying-wing configuration has an 80 cm wingspan. It is built out of Expanded Polypropylene (EPP) with an electric motor mounted at the back and two control surfaces serving as elevons (combined ailerons and elevator
Reply
#2

to get information about the topic "micro air vehicles" full report ppt and related topic refer the page link bellow

http://studentbank.in/report-micro-air-v...e=threaded

http://studentbank.in/report-micro-air-v...ull-report

http://studentbank.in/report-swarming-mi...ode=linear

http://studentbank.in/report-micro-air-v...ode=linear
Reply

Important Note..!

If you are not satisfied with above reply ,..Please

ASK HERE

So that we will collect data for you and will made reply to the request....OR try below "QUICK REPLY" box to add a reply to this page
Popular Searches: micro air vehciles ppt, micro air vehicle pdf, vehichle, design of micro air vehicle, micro air vehicle kits, micro air vehicle abstract, bedienungsanleitung orthofex micro air,

[-]
Quick Reply
Message
Type your reply to this message here.

Image Verification
Please enter the text contained within the image into the text box below it. This process is used to prevent automated spam bots.
Image Verification
(case insensitive)

Possibly Related Threads...
Thread Author Replies Views Last Post
  MICRO CONTROLLER BASED SECURITY SYSTEM USING SONAR ajukrishnan 5 7,146 31-01-2015, 11:55 PM
Last Post: Guest
  8 CANDIDATE QUIZ BUZZER USING 8051 MICRO CONTROLLER (AT89C51) seminar class 3 5,572 10-03-2014, 04:43 PM
Last Post: Guest
  advanced mobile phone signal jammer for gsm cdma and 3g networks with prescheduled ti shilpa16 1 1,678 28-10-2013, 12:17 PM
Last Post: ShayneThill
  wireless sensor networks full report project report tiger 18 16,202 15-07-2013, 12:18 PM
Last Post: computer topic
  IMPLEMENTATION OF SRAM USING MICRO WIND TOOL projectsofme 2 2,710 11-04-2013, 10:53 AM
Last Post: computer topic
  SMS BASED VEHICLE LOCATION FINDER FOR PARKING seminar project explorer 2 2,897 31-01-2013, 02:09 PM
Last Post: seminar details
  DISTRIBUTED CALL ADMISSION CONTROL SCHEME FOR MOBILE MICRO-CELLULAR NETWORKS project topics 2 2,934 24-01-2013, 09:31 PM
Last Post: fake_lover01
  Design & Development of a GSM Based Vehicle Theft Control System seminar class 9 11,418 29-11-2012, 01:15 PM
Last Post: seminar details
  Opportunistic Routing for Wireless Ad Hoc and Sensor Networks: Present and Future seminar presentation 3 2,999 22-11-2012, 01:26 PM
Last Post: seminar details
  fuzzy based call admission control in wideband cdma cellular networks full report project topics 2 2,477 22-11-2012, 01:05 PM
Last Post: seminar details

Forum Jump: