International Journal of Computer Science Trends and Technology (IJCST) – Volume 5 Issue 4, Jul – Aug 2017 ISSN: 2347-8578 www.ijcstjournal.org Page 1 UAV Swarm Co-Ordination and Control Using Grossberg Neural Network Dabbiru Ravi Vikranth Department of Computer Science, B. R. R. Degree College, Visakhapatnam ABSTRACT A UAV swarm (Swarm) has multiple unmanned aerial vehicles (UAV) working in synchronization towards achieving a unified/ universal task. Swarm Co-ordination, primarily deals with communicating with each individual UAV independently from the Ground Control Station (GCS) and negotiation based [8] information exchange between UAVs themselves. The most common type of flight formations are behavioural formations. Leader – follower formation types are the most closely associated with hierarchical class schemes. In the Leader – Follower formation, it is enough to specify a path for the leader to fly as in order to dictate the path taken by formation. The architecture of a system supporting a hierarchical Leader/follower formation scheme for Swarm of Quad copters. One Quad copter acts as a leader, while other acts as follower/ (s). Keywords: — UAV, RNPG Algorithm, Navigation Point, Model Predictive Control, Formation I. SWARM CO-ORDINATION Swarm members work together by communicating their position and other useful information in pre-defined intervals. In order to enable such co-ordination, UAV swarm members need to communicate with each other. In case of dedicated communication infrastructure; swarm itself establishes and maintains an ad-hoc communication network. Communication in Flying Ad-hoc Network (FANET) is focused in UAV – UAV (U2U) and UAV to Infrastructure (U2I) communications. Thus, the degree of mobility of nodes is greater than Mobile Ad-hoc Network (MANET) and Vehicular Ad-hoc Network (VANET). The topology changes frequently (it needs peer to peer networking). The communication range must be greater than other networks. . Ad-hoc On Demand Distance Vector (AODV):- AODV is a reactive protocol, which has same on-demand characteristics like Dynamic Source Routing (DSR) while maintaining different mechanisms of routing table. In AODV, each node stores a routing table, which contains a single record for each destination, while in DSR each node can store multiple entries in its routing table for each destination. In AODV, the source node (and also other relay nodes) stores the next hop information corresponding to each data transmission. AODV routing protocol consists of three phases viz. route discovery, packet transmitting and route maintaining. If the source node has packets to send, it initiates a route discovery process to locate the destination node and then dispatches these packets over a determined route. Discovery process enables determined routes without a loop, and it uses a sequence number to determine an up-to-date route of the destination. An expiration time is used to keep route freshness. In this process, intermediate nodes also update their routing tables. After a route-id is constructed, packets are transferred over it. As a result of mobile nodes, some link failures may occur, and this connection loss triggers a repairing process to maintain the routes. Challenges of FANET: Integrating UAVs into national air space is the need of the hours. This co-ordination will enable the destruction of enemy aircraft with minimal losses. At the same time, these UAVs can be used as electronic jammers and for real time video reconnaissance in enemy areas. Therefore, the collaboration of UAVs and manned air craft should be in networked environment. A FANET uses various wireless communication bands such as VHF, UHF, L-Band, Ku-Band, C-Band etc. These bands are also used in application areas like GSM Networks, Satellite Communications etc. To reduce the frequency congestion related issues, there is a need to standardize these communication bands, signal modulation and multiplexing models. Control of Swarm The term Collision avoidance (CA) [2] represents the scenario where the Quad copter; try to avoid obstacles. Navigation problems of UAVs flying in formation in a free and obstacle laden environments are investigated in this paper. When static obstacles pop-up during the flight, the UAVs are required to turn around them and also avoid collisions between them. In order to achieve these goals, a new dual mode control strategy is proposed i.e. a safe mode is defined as an operation in an obstacle free environment and a danger mode is activated when there is a chance of collision or when there are obstacles in the path. RESEARCH ARTICLE OPEN ACCESS