Abstract:
Mobile robots are increasingly being used for tasks like remote surveillance, sensing and
maintenance. Some of these tasks are critical and require intelligent decision making for
successful completion. It is not always possible to rely exclusively on robot level intelli-
gence to make high impact decisions and hence human supervision is needed during task
execution. To facilitate human-in-the-loop task servicing, the task executing robot is re-
quired to remain connected to a remotely located human operator.
However, robot communication range is typically limited and hence multiple mobile robots
might be deployed to perform the tasks. These robots must coordinate with each other
to dynamically form and maintain a communication link such that network connectivity
exists between the robot servicing the task and the human operator positioned at a sta-
tionary base station.
The development of connectivity aware coordination algorithms is complex due to limited
communication range and presence of obstacles in the search region. In this thesis, we
present a distributed multi-robot algorithm for task servicing with human-in-the-loop con-
straint. Robot control and mission execution is independent of the human operator and
is fully autonomous. The algorithm facilitates indirect collaboration amongst the robotic
agents and uses a combination of graph theoretic and gradient descent based approaches
for path planning. Robots exercise independent decision making on task and role assign-
ment by following a self allocation strategy. This allows dynamic task reassignments and
role exchanges amongst the agents based on increased situational awareness. Our solution
successfully implements obstacle avoidance and deadlock resolution while being scalable
and robust to network and robot failures. To substantiate the claims, we present results
from extensive simulations.
