Flappy Hummingbird: An Open Source Dynamic Simulation of Flapping Wing Robots and Animals

Abstract

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. Flapping Wing Micro Air Vehicles (FWMAVs) hold great promise for closing this performance gap. However, design and control of such systems remain challenging due to various constraints. 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. For simulation validation, we recreated the hummingbird-scale robot developed in our lab in the simulation. System identification was performed to obtain the model parameters. The force generation, openloop and closed-loop dynamic response between simulated and experimental flights were compared and validated. The unsteady aerodynamics and the highly nonlinear flight dynamics present challenging control problems for conventional and learning control algorithms such as Reinforcement Learning. The interface of the simulation is fully compatible with OpenAI Gym environment. As a benchmark study, we present a linear controller for hovering stabilization and a Deep Reinforcement Learning control policy for goal-directed maneuvering. Finally, we demonstrate direct simulation-to-real transfer of both control policies onto the physical robot, further demonstrating the fidelity of the simulation.

A CFD data-driven aerodynamic model for fast and precise prediction of flapping aerodynamics in various flight velocities

“…The CDAM characterized by an analytical expression can be easily combined with the flight dynamics of free-flying insects (Sun et al 2007;Xiong & Sun 2008) and even flapping-wing MAVs to help design flight controllers (Fei et al 2019a). For instance, given the flight dynamics expressed explicitly, it would be easier to derive the control parameters (Yao & Yeo 2019).…”

“…To evaluate the performance of the PD control strategy derived using CDAM, we integrate it into the CFD environment and compare the results with those in the CDAM environment with the governing time-accurate dynamic equations from Sun et al (2007) embedded. We first conduct system identification (Fei et al 2019a) to fine-tune the hovering wing kinematics for CDAM and for CFD. For the default wing motion shown in figure 2, CDAM and CFD produce slightly different values for the forces and torques.…”

Flight Recovery of MAVs with Compromised IMU

“…For the purpose of validation, we deployed our method onto a customized simulation tool [27], which was used to solve the nonlinear dynamics of the vehicle. The physical parameters of the vehicle used in the simulation were obtained by conducting system identification on the test platform.…”

ROSS-LAN: RObotic Sensing Simulation Scheme for Bioinspired Robotic Bird LANding

“…The flexible multibody dynamics of an ornithopter is simulated in [7] considering fluid-structure interaction and flight dynamic behavior to design a robotic model capable of flying in trim. The flappy hummingbird [8] is an open source dynamic simulator of flapping-wing micro aerial vehicles created to facilitate the design and study of the flight control of flapping-wing robots. Although these tools are useful to simulate ornithopters dynamics, kinematics and aerodynamic effects, to the best the author's knowledge there is not a simulation tool designed to retrieve sensor information during their flight.…”