Page 446 - AWSAR 2.0

422 || AWSAR Awarded Popular Science Stories - 2019
Now, the challenge is to place the charging station wisely to meet the above requirement.
To address autonomous recharging, researchers previously proposed the concept of tethering the robot with some uninterrupted power supply. This, however, restricts the robot’s reachable locations heavily; the problem of tangling may also arise. Some research on the docking-based autonomous recharging considered robots covering a workspace with limited battery power. Their objective is to minimize the number of visits to the docking station such that the robot never goes out of charge. Also, research was done where a UAV had multiple re-fuelling depots while performing its designated tasks without failing fuel constraints. Most of the above research considered arbitrary placement of docking stations. In our work, we consider that the robot may reach its power threshold at any location of the workspace, and it should immediately move towards a
charging station nearby to it with
its available charge. Therefore,
the placement of charging
stations can be pivotal to make
the system energy efficient.
Charging station
placement for the robot
inherently depends on the
parameters such as power
threshold, number of charging
stations, locations of the
charging stations and motion
dynamics of the robots. In
this work, we formulate two
variants of the charging station
placement problem. In the first
problem, we assume that the
power threshold for the robot is
given, and we aim to find the minimum number of charging stations and their locations in the workspace. In the second problem, we assume that the number of charging stations
to be installed is given, and we attempt to find their locations and the value of the minimum power threshold for the robot.
We aim to reduce the aforementioned problem to an optimization problem in terms of power threshold and number of charging stations. Two variants of the charging station placement problem are proposed. The first variant assumes a power threshold to be given; we find out a minimum number of charging stations and their locations. The second variant considers the number of charging stations to be given, and we find out the value of the minimum power threshold and charging station locations. Please note that, in both cases, charging stations are placed in a way such that the robot can reach a charging station from any obstacle-free location in the workspace, with the amount of energy remaining with the robot being greater than or equal to the power threshold.
The above problem is reduced to a constraint solving problem. We adopt similar concepts for both the variants of the problem. Let us consider problem variant-1. Decision variables are the number of charging stations and motion plan for the robot. Constraints for the above problem are the workspace, obstacles, robot motion dynamics and power threshold. Using a naive approach, we can simply check for satisfiability of the above problem starting from a small value of charging station count. With every test value of charging station count, we
check for satisfiability of all the constraints. If it is unsatisfiable, check for the immediate next value and so on. Whenever it satisfies all the constraints, it gives a model (solution) that gives
Indoor robots are generally battery-powered, and their battery needs to be recharged at regular intervals. A robot that requires human intervention to constantly monitor its charge level and carry it to the charger is complicated and lacks autonomy, which is fundamental to robotics. Here, we take
a moment to define ‘power threshold’ of the robot.