Autonomous coordinated navigation of drones

TEAM

NamePositionDepartment
Chiranjib Bhattacharyya
(Principal Investigator)
ProfessorRobert Bosch Centre for Cyber-Physical Systems
Computer Science and Automation
S N OmkarPrincipal Research ScientistAerospace Engineering
Arnab BAssistant ProfessorRobert Bosch Centre for Cyber-Physical Systems
Electrical Communication Engineering
Bharadwaj AmruturChairman
Professor
Robert Bosch Centre for Cyber-Physical Systems
Electrical Communications Engineering
Shalabh BhatnagarProfessorRobert Bosch Centre for Cyber-Physical Systems
Computer Science and Automation

Abstract

The broad objective of the project is to achieve autonomous navigation of multiple drones. There are several important problems that need to be answered before we could achieve this objective. We would like to study these questions in the context of Smart Cities, to enable applications  such as delivery of essential supplies like medicines/organs etc. across the city.  

  1. Drone infrastructure in cities: To make widespread usage of drones, viable modern cities need to be adequately equipped with appropriate infrastructure. This is a nascent area which is attracting lot of research interest. It requires dialogue with regulatory authorities and city-planners. Unfortunately to the best of our knowledge there are no design of a suitable infrastructural facility where one can experiment to understand underlying issues related to infrastructural problems in deploying drones. There are two main challenges that need to be addressed namely lack of (1) regulatory guidelines and (2) robust drones which can fly over long distances.
  2. Coordination among multiple drones: It is recognized that coordinating multiple drones is an important open problem. The main challenges are state estimation in the presence of uncertainty and then achieving appropriate coordination without an external command and control structure. Only recently there was some success in creating algorithms which could demonstrate behaviours such as flocking. However much needs to be done and is prompting funding agencies to solicit proposals in coordinating multiple drones. We wish to study the problem of coordination when there is GPS information yielding approximate estimates of state.

Our long-term goals are: (1) Design of drones which can cover city wide areas, (2) suggest infrastructure which could potentially be used in a city for drones, and (3) autonomous navigation strategies for multiple drones.

The goals for the first project year are:  (1) Creating an infrastructure for drones, (2) designing of drones, (3) autonomous navigation of a single drone with GPS, and (4) state perception using computer vision for drones.