Almost all flying creatures, from centimeter-scale insects to meter-scale birds, maneuver through flapping-wing propulsion. Development of flapping-wing robots capable of long-term autonomous flight remains an ongoing challenge owing to manufacturing, actuation, and control considerations. In this article, we report the design and the control system of an all servo-driven bird-like flapping-wing aerial robot called USTBird. Different from most common flapping mechanisms combining gear transmission and motor, USTBird is driven by two independently programmable servos, which makes the wingbeat kinematics of USTBird diverse and enables it to do some agile aerobatic maneuvers. The enhanced controllability brought by the servo-driven method allows USTBird to produce both flight force and control force relying solely on the flapping wings, and exert decoupled control torques about all three body axes without a controllable tail. The flapping gait optimization problem is considered for USTBird, and three different kinds of drive signals are compared for larger aerodynamic forces generation. Toward practical applications, an outdoor flight task of autonomous patrol around a building, which can be subdivided into a coordinated takeoff and a low-altitude patrol and reconnaissance, is conducted on the robot with a good performance.

Flapping-wing robots (FWRs) have great potential in comprehensively optimizing the flexibility, cruising speed, load capacity and the robot mass. Recently, researchers have focused on small-wingspan FWRs but with a large limitation of the payload and endurance. In this paper, we design an eagle-like FWR with a vision system and a flight control system, where the wingspan is as long as 1.78 m and the mass is 985 g (without the vision system). Even though the FWR carries a vision module weighing 165 g, it can cruise for more than 1 hour where the flight speed frequently reaches 5.9 m/s. If the FWR removes the vision system and further adds a 132 g battery, the flight time can be increased to 95 $\sim$ 110 minutes. By considering the vision module as a payload, according to the mass distribution method and successful FWR examples, we roughly determine the mass of the electronic system and FWR structure. The FWR structure is designed to satisfy the mass limitation and most importantly, guarantee the endurance as long as possible. This structure optimization, although is not unique, is of large novelty and different from the formulaic mass distribution method.

Flight testbeds with multiple unmanned aerial vehicles (UAVs) are especially important to support research on multi-vehicle-related algorithms. The existing platforms usually lack a generic and complete solution allowing for software and hardware design. For such a purpose, this paper presents the design and implementation of a comprehensive multi-camera-based testbed for 3-D tracking and control of UAVs. First, the testbed software consists of a multi-camera system and a ground control system, which performs image processing, camera calibration, 3-D reconstruction, pose estimation, and motion control. In the multi-camera system, the positions and orientations of UAVs are first reconstructed by using epipolar geometric constraints and triangulation methods and then filtered by an extended Kalman filter (EKF). In the ground control system, a classical proportional-derivative (PD) controller is designed to receive the navigation data from the multi-camera system and then generates control commands to the target vehicles. Then, the testbed hardware employs smart cameras with field-programmable gate array (FPGA) modules to allow for real-time computation at a frame rate of 100 Hz. Lightweight quadcopter Parrot Bebop drones are chosen as the target UAVs, which does not require any modification to the hardware. Artificial infrared reflective markers are asymmetrically mounted on target vehicles and observed by multiple infrared cameras located around the flight region. Finally, extensive experiments are performed to demonstrate that the proposed testbed is a comprehensive and complete platform with good scalability applicable for research on a variety of advanced guidance, navigation, and control algorithms.

Different from birds, bats, flies, and other flying creatures that generate propulsions by flapping a pair of wings, butterflies with two pairs of wings have their peculiar flight aerodynamics. In this article, a servo-driven biomimetic robotic butterfly, named USTButterfly, is designed to study the flight mechanism of biological butterflies. First, instead of using a single actuator to drive two pairs of wings, two servos are introduced to independently drive the left and right wings, thereby realizing tailless control of the robotic butterfly. Next, a bionic airfoil is designed inspired by the wing structure of the glasswing butterfly. Wing geometry analysis indicates that USTButterfly well matches the morphological characteristics of biological butterflies. Finally, a multicamera motion capture system is used to measure the flight characteristics of the robotic butterfly, and the results demonstrate that despite a larger Reynolds number, USTButterfly shares a similar coupled wing–body interaction with biological butterflies in the climbing flight. USTButterfly opens up a new way to study the flight specializations of biological butterflies, and offers a potential alternative paradigm for existing unmanned aerial vehicles.

This article addresses a novel adaptive compensation strategy for a flexible manipulator in the presence of unmodeled system dynamics and an infinite number of time-varying actuator faults. Differing from the existing results, the unmodeled system dynamics are represented by a nonlinear time-varying delay function, which is tackled with the aid of neural network technique and Lyapunov–Krasovskii function. By introducing the projection technique and a piecewise Lyapunov function, a fault-tolerant control method is proposed, the uniformly ultimate boundedness of the closed-loop system is ensured, and the number of actuator faults is allowed to be infinite. Simulation and physical experiment results show that the manipulator can track the desired angle and the vibration can be suppressed effectively under the proposed scheme for the system with uncertainties and actuator faults.