Benjamin M. Finio

Progress on 'pico' air vehicles

As the characteristic size of a flying robot decreases, the challenges for successful flight revert to basic questions of fabrication, actuation, fluid mechanics, stabilization, and power, whereas such questions have in general been answered for larger aircraft. When developing a flying robot on the scale of a common housefly, all hardware must be developed from scratch as there is nothing ‘off-the-shelf’ which can be used for mechanisms, sensors, or computation that would satisfy the extreme mass and power limitations. This technology void also applies to techniques available for fabrication and assembly of the aeromechanical components: the scale and complexity of the mechanical features requires new ways to design and prototype at scales between macro and microeletromechanical systems, but with rich topologies and material choices one would expect when designing human-scale vehicles. With these challenges in mind, we present progress in the essential technologies for insect-scale robots, or ‘pico’ air vehicles.

Artificial insect wings of diverse morphology for flapping-wing micro air vehicles.

The development of flapping-wing micro air vehicles (MAVs) demands a systematic exploration of the available design space to identify ways in which the unsteady mechanisms governing flapping-wing flight can best be utilized for producing optimal thrust or maneuverability. Mimicking the wing kinematics of biological flight requires examining the potential effects of wing morphology on flight performance, as wings may be specially adapted for flapping flight. For example, insect wings passively deform during flight, leading to instantaneous and potentially unpredictable changes in aerodynamic behavior. Previous studies have postulated various explanations for insect wing complexity, but there lacks a systematic approach for experimentally examining the functional significance of components of wing morphology, and for determining whether or not natural design principles can or should be used for MAVs. In this work, a novel fabrication process to create centimeter-scale wings of great complexity is introduced; via this process, a wing can be fabricated with a large range of desired mechanical and geometric characteristics. We demonstrate the versatility of the process through the creation of planar, insect-like wings with biomimetic venation patterns that approximate the mechanical properties of their natural counterparts under static loads. This process will provide a platform for studies investigating the effects of wing morphology on flight dynamics, which may lead to the design of highly maneuverable and efficient MAVs and insight into the functional morphology of natural wings.

Optimal energy density piezoelectric twisting actuators

The use of piezoelectric materials as actuators or sensors is widespread, and numerous actuator topologies and models have been developed. However, many of these applications do not place stringent requirements on actuator mass or energy density. Motivated by applications that do have strict requirements in these areas such as flapping-wing microrobots, a torsional piezoelectric actuator is developed. A model is presented that predicts output rotation, torque and energy density values; and allows optimization of these values based on actuator geometry. An emphasis is placed on actuator fabrication and testing for empirical validation of the model.

Body torque modulation for a microrobotic fly

The Harvard Microrobotics Lab has previously demonstrated the worlds first at-scale robotic insect capable of vertical takeoff with external power. Both of the robots wings were driven by a single power actuator and 1-DOF mechanical transmission - making independent control of both wings, and therefore asymmetric flapping and the generation of a net body torque, impossible. This paper presents a method to modulate body torques by altering the kinematics of each wing transmission independently, via the introduction of two additional control actuators. Theoretical kinematic and dynamic predictions based on a pseudo-rigid body model are compared to the observed wing trajectories. Controllable body torques are necessary for the development of control algorithms for eventual stable hovering and free flight.

Asymmetric flapping for a robotic fly using a hybrid power-control actuator

This paper continues the exploration of the design space for an insect-sized autonomous flapping-wing MAV with the goal of stable hovering. Previous work has focused on the use of a large primary power actuator to generate flapping motion and smaller “control” actuators to asymmetrically alter wing kinematics. Here a new iteration of this concept is presented, merging the two actuator types to create a “hybrid” power-control actuator. Kinematic and dynamic models for wing motion are presented, and the predictions of these models are compared to experimental results from a prototype design. Controllable asymmetry in wing kinematics can be mapped into controllable body torques via an aerodynamic model, and this information can be used for the generation of control laws for stable hover and eventually highly agile aerial vehicles.

System identification and linear time-invariant modeling of an insect-sized flapping-wing micro air vehicle

Flapping-wing robots typically include numerous nonlinear elements, such as nonlinear geometric and aerodynamic components. For an insect-sized flapping-wing micro air vehicle (FWMAV), we show that a linearized model is sufficient to predict system behavior with reasonable accuracy over a large operating range, not just locally around the linearization state. The theoretical model is verified against an identified model from a prototype robotic fly and implications for vehicle design are discussed.

Open-loop roll, pitch and yaw torques for a robotic bee

This paper presents measurements of open-loop roll, pitch and yaw torques, and open-loop flight experiments for an insect-sized robotic bee. Torques are generated entirely with flapping wings via an actuation scheme that uses a single, central power actuator and two smaller control actuators that fine-tune wing motion. We present an initial 110mg design used for torque measurements and a lighter 83mg prototype that is capable of liftoff with external power and can execute open loop pitching and rolling maneuvers.

Distributed power and control actuation in the thoracic mechanics of a robotic insect

Recent advances in the understanding of biological flight have inspired roboticists to create flapping-wing vehicles on the scale of insects and small birds. While our understanding of the wing kinematics, flight musculature and neuromotor control systems of insects has expanded, in practice it has proven quite difficult to construct an at-scale mechanical device capable of similar flight performance. One of the key challenges is the development of an effective and efficient transmission mechanism to control wing motions. Here we present multiple insect-scale robotic thorax designs capable of producing asymmetric wing kinematics similar to those observed in nature and utilized by dipteran insects to maneuver. Inspired by the thoracic mechanics of dipteran insects, which entail a morphological separation of power and control muscles, these designs show that such distributed actuation can also modulate wing motion in a robotic design.

An ultra-high precision, high bandwidth torque sensor for microrobotics applications

Motivated by the need for torque sensing in the µNm range for experiments with insect-sized flapping-wing robots, we present the design, fabrication and testing of a custom single-axis torque sensor. The micorobots in question are too large for MEMS force/torque sensors used for smaller live insects such as fruit flies, but too small to produce torques within the dynamic range of commercially available force/torque sensors. Our sensor consists of laser-machined Invar sheets that are assembled into a three dimensional beam. A capacitive displacement sensor is used to measure displacement of a target plate when the beam rotates, and the output voltage is correlated to applied torque. Sensor bandwidth, range, and resolution are designed to match the criteria of the robotic fly experiments while remaining insensitive to off-axis loads. We present a final sensor design with a range of ±130µNm, a resolution of 4.5nNm, and bandwidth of 1kHz.

Stroke plane deviation for a microrobotic fly

Wing motion in most flapping-wing micro air vehicles (MAVs) is restricted to a flat stroke plane in order to simplify analysis and mechanism design. An MAV actuation and transmission design capable of controlling flapping motions and deviations from the mean stroke plane using relatively simple modifications to a proven design is presented. This allows preliminary investigation into more power-efficient wing trajectories, an important concern for small MAVs. A theoretical quasi-steady model of flapping wing flight is used to predict wing motions, and these predicted trajectories are compared to empirically observed trajectories from a test device. The ratio of average lift to average aerodynamic power is used as an efficiency metric to compare stroke trajectories.