Chapter

Cooperative Robots and Sensor Networks 2014

Volume 554 of the series Studies in Computational Intelligence pp 87-113

Collaborative Autonomous Surveys in Marine Environments Affected by Oil Spills

  • Shayok MukhopadhyayAffiliated withSchool of Electrical and Computer Engineering, Georgia Institute of Technology Email author 
  • , Chuanfeng WangAffiliated withGeorge W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology
  • , Mark PattersonAffiliated withMarine Science Center, Northeastern University
  • , Michael MalisoffAffiliated withDepartment of Mathematics, Louisiana State University
  • , Fumin ZhangAffiliated withSchool of Electrical and Computer Engineering, Georgia Institute of Technology

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This chapter presents results on collaborative autonomous surveys using a fleet of heterogeneous autonomous robotic vehicles in marine environments affected by oil spills. The methods used for the surveys are based on a class of path following controllers with mathematically proven convergence and robustness. Use of such controllers enables easy mission planning for autonomous marine surveys where the paths consist of lines and curves. The control algorithm uses simple dynamic models and simple control laws and thus enables quick deployment of a fleet of autonomous vehicles to collaboratively survey large areas. This enables using a mobile network to survey an area where the different member nodes may have slightly different capabilities. A mapping algorithm used to reconcile data from heterogeneous marine vehicles on multiple different paths is also presented. Vehicles with heterogeneous dynamics are thus used to aid in the reconstruction of a time varying field. The algorithms used were tested, mainly on student-built marine robots that collaboratively surveyed a coastal lagoon in Grand Isle, Louisiana that was polluted by crude oil during the Deepwater Horizon oil spill. The results obtained from these experiments show the effectiveness of the proposed methods for oil spill surveys and also provide guidance for mission designs for future collaborative autonomous environmental surveys.