BPACAR: design of a hybrid bioinspired model for dynamic collision-aware routing with continuous pattern analysis in UAV networks

Designing collision-aware routing (path planning) protocols for UAV (Unmanned Aerial Vehicle) Networks requires multimodal analysis of various network and node-level parameter sets. These include node-to-node distance, energy constraints, communication constraints, QoS (Quality of Service) constraints, etc. Existing collision-aware UAV routing models are either highly complex or have lower efficiency, which limits their deployment abilities. Moreover, these models usually do not consider energy constraints and are applied to static targets. To overcome these limitations, this article gave an idea about the design of a novel hybrid bioinspired model. The proposed model initially collects node-level and network-level parametric sets that include Cartesian location, residual energy levels, temporal routing performance, and temporal collision performance levels. The model then deploys a Grey Wolf Optimization (GWO) based routing process to identify optimal routes between two anchor points. The routes are again tuned via a Firefly based Optimization (FFO) which assists in estimating high-trust routes based on their temporal performance via continuous data update operations. The selected route sets are further scrutinized via a continuous learning framework (CLF), which assists in the identification of dynamic moving targets, and uses this information for incremental route updates. Due to the integration of CLF, the model can identify optimal paths even under moving target scenarios. The model was validated under multiscale networks, and its performance was evaluated in terms of collision avoidance accuracy, routing delay, energy requirements, and computational complexity levels w.r.t. dynamic scenarios. This performance was compared with various state-of-the-art methods, and it was seen that the proposed model has 10.5% lower routing delay, with 8.3% lower energy consumption, and 23.9% lower collisions while maintaining lower computational complexity. Due to these enhancements, the proposed model can deploy a wide variety of real-time UAV network scenarios.