Building control laws for multi-agent systems

Imagine you are driving to work. You know that at a constant speed of 50 kmph along the pre-determined path, it will take an hour to reach office. On the way, you are stuck in a traffic jam for 15 minutes. You now need to incorporate this information (feedback) and take corrective action by increasing your speed (control) in order to reach on time. Jerome Moses, a research scholar with the IITB-Monash Research Academy, realised that feedback and control are inherent in almost everything we do, from singing to driving to playing football. This is what prompted him to take up a project titled, ‘Distributed coordination of mobile visual sensor network for boundary tracking and coverage applications’ for his PhD.

Figure 1. An AR.Drone 2.0 (Quadcopter), which is used to test the cooperative control algorithms built. (Courtesy: parrot.com)

The Academy is a collaboration between India and Australia that endeavours to strengthen scientific relationships between the two countries. Graduate research scholars like Jerome study for a dually-badged PhD from both IIT Bombay and Monash University, spending time at both institutions to enrich their research experience. “I am attempting to build cooperative control laws / algorithms that will help multi-agent systems like unmanned air-vehicles or wheeled-mobile robots to achieve a particular task. The tasks can vary from covering a given area or conducting surveillance of a specified zone to tracking a boundary,” explains Jerome.

Figure 2. Top view of five mobile robots (marked 1 to 5) trying to achieve a formation. The patterned markers are used for identification/localisation of the mobile robots.

Since Jerome plans to implement his research on a robotic-test bed, there are many practical constraints and realistic issues that need to be addressed. These involve communication delays between the agents, loss of information, sensing, and actuating constraints of the input to these agents. However, the benefits of my research are tremendous, says Jerome. “For instance, the coordination laws I build can be used to cover an area in a repeated manner. Area coverage essentially requires two things, (i) every point in the area needs to be covered, and (ii) the time of revisit of a particular point needs to be ensured. Another application of the algorithms is covering the boundary of a given zone in a coordinated way.” Says Prof Murali Sastry, CEO of the IITB-Monash Research Academy, “Jerome’s research has immense applications — from target tracking, monitoring poisonous oil spills, harmful algae blooms, temperature and salinity distribution in the ocean, area coverage, surveillance, patrolling, and even rescue missions. The Academy was conceived as a unique model for how two leading, globally focused academic organisations can come together in the spirit of collaboration to deliver solutions and outcomes to grand challenge research questions facing industry and society. We are confident that scholars like Jerome Moses will do us proud.”

Figure 3. Possible trajectories that can be generated with control laws (from left: rendezvous, trochoid, spiral, circle)

Research scholar: Jerome Moses , IITB-Monash Research Academy

Project title: Distributed coordination of mobile visual sensor network for boundary tracking and coverage applications

Supervisors: Prof. Arpita Sinha, Prof. Hoam Chung

Contact details: jeromemoses@iitb.ac.in

This story was written by Mr Krishna Warrier based on inputs from the research student, his supervisors and IITB-Monash Research Academy. Copyright IITB-Monash Research Academy.