A new metaheuristic is proposed for optimisation of unconstrained continuous domain cost functions, taking inspiration in the foraging behaviour of populations of urban pigeons in the public parks of large cities. We discuss the ideas underlying the computational model of the resulting optimiser, particularly its exploration vs exploitation strategies, and verified the applicability of the algorithm on a number of benchmark functions, obtaining error deviations from the optimum smaller than 0.1%. Despite the simplicity of the model, its effectiveness is comparable to other recently proposed swarm bird-inspired techniques. Besides, we discuss ideas to enhance the model towards the context of non-stationary function optimisation, a natural challenge pigeons undertake during their daily-basis food discovering journeys.