Routing protocols for Mobile Ad-hoc NETworks (MANETs) have been explored extensively in recent years. Amongst these routing protocols, Ad hoc On-demand Distance Vector (AODV) protocol is one of the reactive unicast routing protocol for mobile ad hoc networks. This paper proposes a Radial Basis Function Based Neural Network Approach for Ad-hoc On-demand Distance Vector Protocol for MANETs (RBFNN based AODV). This protocol imports the intelligence of RBF neural network function into the existing AODV protocol. The RBFNN based AODV allows each node in the network to send out unicast data packets intelligently. The authors' proposed scheme has node to node data transmission ongoing more than one mobile nodes using random mobility functions on and off.