CPG-based neural control for a flapping-wing ornithopter

With the increasing importance of drones and the expansion of basic technological capabilities, the development of flying robots based on biomimetic principles has become more relevant. Birds are capable of flying in obstacle-rich environments such as forests, recovering flight after collisions, and transitioning to energy-efficient gliding modes by utilizing wind currents to maintain flight with minimal energy expenditure. However, to achieve similar advantages, an ornithopter must accurately mimic the wing kinematics of its biological prototype and consequently possess a biologically similar control system. Neural control in flapping-wing robotic systems allows for adaptive and precise tuning of biomimetic motion patterns and may potentially enable the use of evolutionarily refined flight techniques. Our research focuses on developing a neural network model to generate oscillatory movements in the actuation system of a biomimetic flapping-wing robot.