Autonomous UAVs for urban traffic monitoring must respond quickly to changing operational conditions while maintaining stable, transparent decision-making. Rule-based controllers respond only at predefined thresholds, while learning-based methods adapt well but lack the certification transparency required for safety-critical deployment. This paper proposes a bio-inspired emotion-regulated decision-control mechanism and introduces the Pull–Push Engine (PPE), a regulatory architecture that balances environmental stimuli against personality-anchored baselines through weighted temporal integration. The PPE is embedded in a three-layer framework combining Big Five personality traits, the Pleasure–Arousal–Dominance (PAD) model, and Ortony–Clore–Collins (OCC) event appraisal. Validation in a SUMO-based simulation across three scenarios of increasing complexity showed that PPE regulation maintained bounded PAD trajectories and zero saturation despite concurrent stressors, whereas removing the pull term caused 57–88% saturation. Behavioral diversity scaled naturally with operational demands: Surprised mood dominated across all scenarios (47.8–67.5%), with Anxious and Focused increasing systematically with complexity. Strategy entropy rose monotonically (1.885–2.033 bits). A sensitivity sweep confirmed robust regulation across a stable operating region, with degradation only at the boundary (p < 0.001 for all key comparisons). Every simulated decision remains causally traceable from stimulus through emotional processing to action. This ensures interpretability, which is essential for future safety-critical UAV deployment, although hardware implementation and field validation are still required.
Zaidan et al. (Sun,) studied this question.