Enforcing nonholonomic constraints in Aerobat, a roosting flapping wing model
This work addresses the challenge of biomimetic robotics for flapping wing flight, specifically for bat-like robots, but appears incremental as it builds on existing control methods for articulated systems.
The paper tackled the problem of controlling a bio-inspired bat robot (Aerobat) for flapping flight by deriving a multi-body dynamical system and enforcing nonholonomic constraints to achieve closed-loop aerial body reorientation and landing preparation, resulting in a model and optimization-based nonlinear controller that resolves joint trajectories to achieve desired angular momentum about the roll axis.
Flapping wing flight is a challenging dynamical problem and is also a very fascinating subject to study in the field of biomimetic robotics. A Bat, in particular, has a very articulated armwing mechanism with high degrees-of-freedom and flexibility which allows the animal to perform highly dynamic and complex maneuvers, such as upside-down perching. This paper presents the derivation of a multi-body dynamical system of a bio-inspired bat robot called Aerobat which captures multiple biologically meaningful degrees-of-freedom for flapping flight that is present in biological bats. Then, the work attempts to manifest closed-loop aerial body reorientation and preparation for landing through the manipulation of inertial dynamics and aerodynamics by enforcing nonholonomic constraints onto the system. The proposed design paradigm assumes for rapidly exponentially stable controllers that enforce holonomic constraints in the joint space of the model. A model and optimization-based nonlinear controller is applied to resolve the joint trajectories such that the desired angular momentum about the roll axis is achieved.