This work aims to develop an integrated conceptual design process to assess the scalability and performance of propulsion systems of resonant motor-driven flapping wing vehicles. Direct simulation-to-real transfer is achieved, demonstrating the hummingbird-like fast evasive maneuvers on the at-scale hummingbird robot. The hybrid policy manifests a maneuver that is close to that observed in hummingbirds. A model-free reinforcement learning policy trained in simulation was optimized to 'destabilize' the system and maximize the performance during maneuvering. However, during extreme maneuver, the modeling error becomes unmanageable. We use model-based nonlinear control for nominal flight control, as the dynamic model is relatively accurate for these conditions. The proposed hybrid control policy combines model-based nonlinear control with model-free reinforcement learning. Inspired by the hummingbirds' near-maximal performance during such extreme maneuvers, we developed a flight control strategy and experimentally demonstrated that such maneuverability can be achieved by an at-scale 12-gram hummingbird robot equipped with just two actuators. Consider the wingbeat frequency of 40Hz, this aggressive maneuver is carried out in just 0.2 seconds. Given a sudden looming visual stimulus at hover, a hummingbird initiates a fast backward translation coupled with a 180-degree yaw turn, which is followed by instant posture stabilization in just under 10 wingbeats. Finally, we demonstrate direct simulation-to-real transfer of both control policies onto the physical robot, further demonstrating the fidelity of the simulation.īiological studies show that hummingbirds can perform extreme aerobatic maneuvers during fast escape. As a benchmark study, we present a linear controller for hovering stabilization and a Deep Reinforcement Learning control policy for goal-directed maneuvering. The interface of the simulation is fully compatible with OpenAI Gym environment. The unsteady aerodynamics and the highly nonlinear flight dynamics present challenging control problems for conventional and learning control algorithms such as Reinforcement Learning. The force generation, open-loop and closed-loop dynamic response between simulated and experimental flights were compared and validated. System identification was performed to obtain the model parameters. For simulation validation, we recreated the hummingbird-scale robot developed in our lab in the simulation. Here, we present an open source high fidelity dynamic simulation for FWMAVs to serve as a testbed for the design, optimization and flight control of FWMAVs. However, design and control of such systems remain challenging due to various constraints. Flapping Wing Micro Air Vehicles (FWMAVs) hold great promise for closing this performance gap. Insects and hummingbirds exhibit extraordinary flight capabilities and can simultaneously master seemingly conflicting goals: stable hovering and aggressive maneuvering, unmatched by small scale man-made vehicles. In addition, further integration with other modes of locomotion, such as crawling, jumping, perching, self-wing-folding, and water-diving, can be a future direction of a FWAV to fully adapt the biologically locomotive strategies in nature, and to increase the range of applications. Achievements in the development of FWAVs demonstrate their potential for future applications, both in the military and civilian fields. We discuss the capability of free flight and flight endurance of the FWAVs, which are limited by current electronics and power technologies that severely constrain those vehicles using other driving actuators, rather than conventional electromagnetic motors, to freely take off. In this review paper, we survey recent developments of insect-inspired tailless FWAVs in various sizes from micro- to pico-scale, with different types of driving actuator, mechanism design, wing configuration, and control strategy. Consequently, among the many developed vehicles, only a few are capable of free flight. However, the technical difficulty involved in designing and building such a complicated and compact system within a limited takeoff weight for it to remain airborne is a major barrier. Insects have therefore become a source of inspiration for the development of tailless, hover-capable flapping-wing air vehicles (FWAVs). Flying insects are able to hover and perform agile maneuvers by relying on their flapping wings to produce control forces, as well as flight forces, due to the absence of tail control surfaces.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |