Collision avoidance between UAV clusters using swarm intelligence techniques

Abstract

In this article, several basic swarming laws for Unmanned Aerial Vehicles (UAVs) are developed for both twodimensional (2D) plane and three-dimensional (3D) space. Effects of these basic laws on the group behaviour of swarms of UAVs are studied. It is shown that when cohesion rule is applied an equilibrium condition is reached in which all the UAVs settle at the same altitude on a circle of constant radius. It is also proved analytically that this equilibrium condition is stable for all values of velocity and acceleration. A decentralised autonomous decision-making approach that achieves collision avoidance without any central authority is also proposed in this article. Algorithms are developed with the help of these swarming laws for two types of collision avoidance, Group-wise and Individual, in 2D plane and 3D space. Effect of various parameters are studied on both types of collision avoidance schemes through extensive simulations.

Bio-inspired group mobility model for mobile ad hoc networks based on bird-flocking behavior

“…The nodes in our model are not guided by a centralized entity; rather they use locally available information to route their movement. Basic swarming laws such as cohesion, following, homing, dispersion and alignment (Sharma and Ghose 2009) have been used to model the movement of independent agents (hereafter, referred to as nodes). At each time interval, the direction of movement of nodes is decided by the combined interaction of these basic forces.…”

System for deployment of groups of unmanned micro aerial vehicles in GPS-denied environments using onboard visual relative localization

“…Recent research on multi-MAV systems has focused on aspects of communication and maintenance of connectivity within the team members [2], [3], modeling of the swarm behavior by predicting individual behaviors [4], [5], task allocation and strategies for solving multiple tasks [6], [7], [8], and control and collision avoidance within the swarm [9], [10], [11], [12]. Topics covered in this paper are related mainly to control and stabilization of MAV teams.…”

Underwater Robot Swarms: Challenges and Opportunities

“…The emergent equilibrium in the system allows the robots to avoid collisions or large separations, while having the group advance in a common direction. Several researchers have found that only implementing "separation" and either "alignment" or "cohesion" is often sufficient to achieve coherent flocking (De Nardi, 2004;Park et al, 2003;Sharma and Ghose, 2009). Moreover, work by Kadrovach and Lamont (2001) showed that applying "separation" and "cohesion" to robots with no default forward speed (rotor-crafts) could lead to the formation of a stable grid which could be used to cover an area and serve as a sensor network in the sky.…”