Modeling and control of flapping wing micro aerial vehicles

Abstract

Research in robots that emulate insect flight or micro aerial vehicles (MAV) has gained significant momentum in the past decade owing to the vast number of fields they could be employed in. In this paper, key modeling and control aspects of a flapping wing MAV in hover have been discussed. Models of varying complexity have been developed by previous researchers. Here, we examine the validity of key assumptions involved in some of these models in a closed-loop control setting. Every model has limitations and with proper design of feedback control these limitations can be overcome up to a certain degree. Three nonlinear models with increasing complexity have been developed. Model I includes only the rigid body dynamics while ignoring the wing dynamics while model II includes the rigid body dynamics along with the wing kinematics. Lastly, model III encompasses the complete rigid body and the rigid wing dynamics. To ensure these higher fidelity models can be rendered unnecessary with a suitably designed controller, a method is presented wherein the controller is designed for the simplest model and tested for its robustness on the more complex models. Linear quadratic regulator (LQR) is used as the main control system design methodology. A nonlinear parameter optimization algorithm is employed to design a family of LQR control systems for the MAV. Additionally, critical performance trade-offs are illuminated, and properties at both the plant output and input are examined. Lastly, we also provide specific rules of thumb for the control system design.

Full Flight Envelope and Trim Map of Flapping-Wing Micro Aerial Vehicles

“…where x ∈ R n is the state, u ∈ R m is the control input, p ∈ R l is a vector of parameters, and x 0 is the initial condition [15][16][17][18][19][20][21][23][24][25][38][39][40][41]51,52,65,66]. A key feature that led to some of the first successful FWMAV flights is that the robot is minimally actuated to meet stringent size and weight constraints [26,39].…”

“…Although many flapping flight controllers have been proposed in the literature [13][14][15][16][17][18][19][20][21][22][23][24], to date control designs have been demonstrated primarily for hovering regimes and symmetric longitudinal flight [24,25]. The reason is that modeling and analyzing the robot dynamics away from these regimes is made difficult by the nonlinear, periodic, and time-varying nature of asymmetric flapping flight [26].…”

“…Although many flapping-wing flight models have been proposed in the literature [16,23,24,[37][38][39][40][41][42][43][44][45][46][47][48][49], methods for capturing the robot six-degree-of-freedom dynamics typically assume negligible wing inertia [15,16,39]. Other methods derive new aerodynamic forces and moments that capture unsteady flow effects on the wings by assuming symmetric flapping and pure longitudinal or hovering flight [19,24,25,42,47,[50][51][52][53].…”

LPV Modeling and Robust Sampled-Data H∞ Control of a Tailless Flapping Wing Micro Aerial Vehicle With Parameter Uncertainties

“…The lack of complete understanding of the nonlinear timevarying dynamics and appropriate simplification leads to an inability to get elegant and accurate models for designing controllers [10]. Biswal et al proposed that the absence of wing inertia has no detrimental effect on controller design, and it is feasible to substitute a time-invariant model for a timevarying system by averaging the period [11]. Flapping counter force (FCF) and Flapping counter torque (FCT), which are proportional to body velocity and wing beat frequency, have been proposed to represent flapping aerodynamic counterparts [12].…”

Multi-cell energy-absorbing structures with hollow columns inspired by the beetle elytra

“…Although the design and investigation of the mechanics of thin-walled structures has been quite popular in recent years, a relatively less explored field is the development of bionic lightweight structures. From the 4D printing of structural elements designed from bio-inspired botanical systems [32] to the design of micro air vehicles biomimicking the fruit fly [33], nature has had a profound impact on the development of modern science and technology. In addition to theoretical and experimental activities [34], finite element simulation has been widely used to explore the mechanical characteristics of bionic structures [35][36][37][38].…”