A Comparative Analysis of Mobile Robot Navigation Strategies

Aljeroudy, Yazan, Kevin Barry, Chaomin Lou, Mohan Krishnan, and Mark Paulik

Robots are increasingly being used or considered for a wide variety of civilian and military applications such as fire fighting, enforcing perimeter security, bomb disposal, exploration, etc. In this context, robot navigation in what may be wholly or partially unknown environments is a challenging problem. It typically requires satisfying a number of different behaviors such as obstacle avoidance, goal finding, trap avoidance, etc., all in real time. This is achieved using sensor information in a reactive multi-behavior-fusion strategy, which produces the right navigation decision by combining appropriate behaviors as needed for any situation the robot might find itself in.

VFH, VFH+, VPH and VPH+ are different navigation algorithms developed by various researchers in the field. In their different ways they extract high-level information from the raw sensor data using which navigation decisions are made. Different navigation algorithms have their individual advantages and disadvantages. Tradeoffs characterize factors such as how effectively they perform in various environmental situations, how fast these algorithms run, etc. In this work a comparative performance analysis is carried out of the effectiveness of these algorithms in various robot environments on the basis of three factors – data reduction, computation reduction, and a cost function.