Single-loop control and trajectory following of a flapping-wing microrobot

Abstract

Abstract-Inspired by the agility of flying insects and the recent development on an insect-scale aerial vehicle, we propose a single-loop adaptive flight control suite designed with an emphasis on the ability to track dynamic trajectories as a step towards the goal of performing acrobatic maneuvers as observed in real insects. Instead of the conventional approach of having cascaded control loops, the proposed controller directly regulates the commanded torques to stabilize the attitude and lateral position in a single loop. One of the recent successful prototypes from the RoboBees project demonstrated its first unconstrained flight [6]. The 80mg Micro Aerial Vehicle (MAV) shown in figure 1 is a result of the culmination of research in meso-scale actuation and manufacturing technology [7], [8]. The flappingwing robot is able to generate body torques and sufficient lift force [9] satisfying the key requirements for stable flight with the aid of an active flight controller [6].In an effort to improve the flight performance demonstrated in [6], the lack of comprehensive knowledge of the system and variation caused by imperfect fabrication motivated the development of a suite of adaptive flight controller capable of coping with model uncertainties [10]. This brought about marked improvement in flight performance as evidenced by a reduction in position errors, particularly for hovering flights. Moreover, the fidelity of this flight controller was further demonstrated in vertical takeoff and landing flights.

Perching with a robotic insect using adaptive tracking control and iterative learning control

“…The scheme benefits from ease of implementation, guaranteed convergence, and compatibility with the vehicle’s nonlinear dynamics, compared to alternatives such as L 1 -adaptive control that requires linearized dynamics (Guerreiro et al, 2009; Ioannou et al, 2014) or a neural-based adaptive method that necessitates a training period (Coza et al, 2011; Madani and Benallegue, 2008; Nicol et al, 2011). In this section, we present the derivation of the proposed altitude controller and the trajectory tracking controller, briefly introduced in Chirarattananon et al. (2014c), based on techniques borrowed from the sliding mode control method (Bouabdallah and Siegwart, 2005; Slotine et al, 1991).…”

“…Here we present one possible implementation of heading control not previously discussed in Chirarattananon et al (2014c). Consider the scenario where we would like the i ^ -axis of the robot to point in the direction of the vector I ^ h defined on the I ^ − J ^ plane in the inertial frame.…”

“…In Section 3, we provide the derivation of the proposed adaptive tracking controller suitable for the flapping-wing robot and other classes of MAVs with similar dynamics, including quadrotors. The controller was first presented with limited experimental results in Chirarattananon et al (2014c). This controller allows the vehicle to track trajectories in the SE ( 3 ) space with provable asymptotic stability under some reasonable assumptions.…”

An at-scale tailless flapping wing hummingbird robot: II. Flight control in hovering and trajectory tracking

“…Extensive research has been conducted to the controller design [20][21][22][23][24][25][26], where averaging theory is commonly employed to allow approximation of the time-varying system dynamics with its time-invariant average [27,28]. For flapping-wing micro aerial vehicles (FWMAVs), provided that the flapping frequencies are sufficiently high, a controller can be designed on the linearized averaged system where the aerodynamic forces are averaged over one or several wingbeats [17,19,20,[29][30][31]. In our previous work, a geometric controller with global exponential attractiveness for attitude control is designed to address system nonlinearity [19].…”