Attitude Stabilization of a Novel Flying Robot by Dynamic Compensation

This paper gives details about the controller design that aims to stabilize the novel twinrotor flying robot, Toruk. Toruk is an experimental test bench to study center of gravity steering, effect of the location of the center of gravity, controller design and implementation, etc. Physical components are also briefly discussed in this paper. Attitude dynamics of the system is inherently unstable. It is stabilized by a regulator. In addition, an observer is designed and utilized to estimate the unmeasured states. Thrust force generated by the propulsion unit is estimated by using the identified mathematical model of the unit. An experimental setup is employed to identify the mathematical model that expresses the relation between the applied input voltage to the propulsion unit and thrust produced by the propeller. Mathematical model for the attitude dynamics of Toruk is built. Then controllability and observability analysis are carried out for the system. Dynamic compensator composed of a state observer and a regulator, is designed on the mathematical model. Physical implementation on the system will be performed.

Autonomous Formation Flight Control of Large-Sized Flapping-Wing Flying Robots Based on Leader–Follower Strategy

Birds in nature exhibit excellent long-distance flight capabilities through formation flight,which could reduce energy consumption and improve flight efficiency.Inspired by the biological habits of birds,this paper proposes an autonomous formation flight control method for Large-sized Flapping-Wing Flying Robots(LFWFRs),which can enhance their search range and flight efficiency.First,the kinematics model for LFWFRs is established.Then,an autonomous flight controller based on this model is designed,which has multiple flight control modes,including attitude stabilization,course keeping,hovering,and so on.Second,a formation flight control method is proposed based on the leader–follower strategy and periodic characteristics of flapping-wing flight.The up and down fluctuation of the fuselage of each LFWFR during wing flapping is considered in the control algorithm to keep the relative distance,which overcomes the trajectory divergence caused by sensor delay and fuselage fluctuation.Third,typical formation flight modes are realized,including straight formation,circular formation,and switching formation.Finally,the outdoor formation flight experiment is carried out,and the proposed autonomous formation flight control method is verified in real environment.

Autonomous flight control with different strategies applied during the complete flight cycle for flapping-wing flying robots

Flapping-wing flying robots(FWFRs),especially large-scale robots,have unique advantages in flight efficiency,load capacity,and bionic hiding.Therefore,they have significant potential in environmental detection,disaster rescue,and anti-terrorism explosion monitoring.However,at present,most FWFRs are operated manually.Some have a certain autonomous ability limited to the cruise stage but not the complete flight cycle.These factors make an FWFR unable to give full play to the advantages of flapping-wing flight to perform autonomous flight tasks.This paper proposed an autonomous flight control method for FWFRs covering the complete process,including the takeoff,cruise,and landing stages.First,the flight characteristics of the mechanical structure of the robot are analyzed.Then,dedicated control strategies are designed following the different control requirements of the defined stages.Furthermore,a hybrid control law is presented by combining different control strategies and objectives.Finally,the proposed method and system are validated through outdoor flight experiments of the HIT-Hawk with a wingspan of 2.3 m,in which the control algorithm is integrated with an onboard embedded controller.The experimental results show that this robot can fly autonomously during the complete flight cycle.The mean value and root mean square(RMS)of the control error are less than 0.8409 and 3.054 m,respectively,when it flies around a circle in an annular area with a radius of 25 m and a width of 10 m.

A novel autonomous self-assembly distributed swarm flying robot

Swarm intelligence embodied by many species such as ants and bees has inspired scholars in swarm robotic researches. This paper presents a novel autonomous self-assembly distributed swarm flying robot-DSFR, which can drive on the ground, autonomously accomplish self-assembly and then fly in the air coordinately. Mechanical and electrical designs of a DSFR module, as well as the kinematics and dynamics analysis, are specifically investigated. Meanwhile, this paper brings forward a generalized adjacency matrix to describe configurations of DSFR structures. Also, the distributed flight control model is established for vertical taking-off and horizontal hovering, which can be applied to control of DSFR systems with arbitrary configurations. Finally, some experiments are carried out to testify and validate the DSFR design, the autonomous self-assembly strategy and the distributed flight control laws.

Extended State Observer based Attitude Control of a Bird-like Flapping-wing Flying Robot

The attitude control system of a flapping-wing flying robot plays an important role in the precise orientation and tracking of the robot.In this paper,the modeling of a bird-like micro flapping-wing system is introduced,and the design of a sliding mode controller based on an Extended State Observer(ESO)is described.The main design difficulties are the control law and the adaptive law for the attitude control system.To address this problem,a sliding mode adaptive extended state observer algorithm is proposed.Firstly,a new extended state approximation method is used to estimate the final output as a disturbance state.Then,a sliding mode observer with good robustness to the model approximation error and external disturbance is used to estimate the system state.Compared with traditional algorithms,this method is not only suitable for more general cases,but also effectively reduces the influence of the approximation error and interference.Next,the simulation and experiment example is given to illustrate the implementation process.The results show that the algorithm can effectively estimate the state of the attitude control system of the flapping-wing flying robot,and further guarantee the robustness of the model regarding error and external disturbance.

Concrete Defects Inspection and 3D Mapping Using City Flyer Quadrotor Robot

The concrete aging problem has gained more attention in recent years as more bridges and tunnels in the United States lack proper maintenance. Though the Federal Highway Administration requires these public concrete structures to be inspected regularly, on-site manual inspection by human operators is time-consuming and labor-intensive. Conventional inspection approaches for concrete inspection, using RGB imagebased thresholding methods, are not able to determine metric information as well as accurate location information for assessed defects for conditions. To address this challenge, we propose a deep neural network(DNN) based concrete inspection system using a quadrotor flying robot(referred to as City Flyer) mounted with an RGB-D camera. The inspection system introduces several novel modules. Firstly, a visual-inertial fusion approach is introduced to perform camera and robot positioning and structure 3 D metric reconstruction. The reconstructed map is used to retrieve the location and metric information of the defects.Secondly, we introduce a DNN model, namely Ada Net, to detect concrete spalling and cracking, with the capability of maintaining robustness under various distances between the camera and concrete surface. In order to train the model, we craft a new dataset, i.e., the concrete structure spalling and cracking(CSSC)dataset, which is released publicly to the research community.Finally, we introduce a 3 D semantic mapping method using the annotated framework to reconstruct the concrete structure for visualization. We performed comparative studies and demonstrated that our Ada Net can achieve 8.41% higher detection accuracy than Res Nets and VGGs. Moreover, we conducted five field tests, of which three are manual hand-held tests and two are drone-based field tests. These results indicate that our system is capable of performing metric field inspection,and can serve as an effective tool for civil engineers.

Research on Gliding Aerodynamic Effect of Deformable Membrane Wing for a Robotic Flying Squirrel

Inspired by creatures with membrane to obtain ultra-high gliding ability, this paper presents a robotic flying squirrel （a novel gliding robot） characterized as membrane wing and active membrane deformation. For deep understanding of membrane wing and gliding mechanism from a robotic system perspective, a simplified blocking aerodynamic model of the deformable membrane wing and CFD simulation are finished. In addition, a physical prototype is developed and wind tunnel experiments are carried out. The results show that the proposed membrane wing is able to support the gliding action of the robot. Meanwhile, factors including geometry characteristics, material property and wind speed are considered in the experiments to investigate the aerodynamic effects of the deformable membrane wing deeply. As a typical characteristic of robotic flying squirrel, deformation modes of the membrane wing not only affect the gliding ability, but also directly determine the effects of the posture adjustment. Moreover, different deformation modes of membrane wing are illustrated to explore the possible effects of active membrane deformation on the gliding performance. The results indicate that the deformation modes have a significant impact on posture adjustment, which reinforces the rationality of flying squirrel＇s gliding strategy and provides valuable information on prototype optimal design and control strategy in the actual gliding process.

ATI: Assemble topological interaction overcomes consistency– cohesion trade‐off in bird flocking

In nature,various animal groups like bird flocks display proficient collective navigation achieved by maintaining high consistency and cohesion simultaneously.Both metric and topological interactions have been explored to ensure high consistency among groups.The topological interactions found in bird flocks are more cohesive than metric in-teractions against external perturbations,especially the spatially balanced topological interaction(SBTI).However,it is revealed that in complex environments,pursuing cohesion via existing interactions compromises consistency.The authors introduce an innovative solution,assemble topological interaction,to address this challenge.Con-trasting with static interaction rules,the new interaction empowers individuals with self-awareness to adapt to the complex environment by switching between interactions through visual cues.Most individuals employ high-consistency k-nearest topological interaction when not facing splitting threats.In the presence of such threats,some switch to the high-cohesion SBTI to avert splitting.The assemble topological interaction thus transcends the limit of the trade-off between consistency and cohesion.In addition,by comparing groups with varying degrees of these two features,the authors demonstrate that group effects are vital for efficient navigation led by a minority of informed agents.Finally,the real-world drone-swarm experiments validate the applicability of the proposed interaction to artificial robotic collectives.

An audio‐based risky flight detection framework for quadrotors

Drones have increasingly collaborated with human workers in some workspaces,such as warehouses.The failure of a drone flight may bring potential risks to human beings'life safety during some aerial tasks.One of the most common flight failures is triggered by damaged propellers.To quickly detect physical damage to propellers,recognise risky flights,and provide early warnings to surrounding human workers,a new and compre-hensive fault diagnosis framework is presented that uses only the audio caused by pro-peller rotation without accessing any flight data.The diagnosis framework includes three components:leverage convolutional neural networks,transfer learning,and Bayesian optimisation.Particularly,the audio signal from an actual flight is collected and trans-ferred into time–frequency spectrograms.First,a convolutional neural network‐based diagnosis model that utilises these spectrograms is developed to identify whether there is any broken propeller involved in a specific drone flight.Additionally,the authors employ Monte Carlo dropout sampling to obtain the inconsistency of diagnostic results and compute the mean probability score vector's entropy(uncertainty)as another factor to diagnose the drone flight.Next,to reduce data dependence on different drone types,the convolutional neural network‐based diagnosis model is further augmented by transfer learning.That is,the knowledge of a well‐trained diagnosis model is refined by using a small set of data from a different drone.The modified diagnosis model has the ability to detect the broken propeller of the second drone.Thirdly,to reduce the hyperparameters'tuning efforts and reinforce the robustness of the network,Bayesian optimisation takes advantage of the observed diagnosis model performances to construct a Gaussian pro-cess model that allows the acquisition function to choose the optimal network hyper-parameters.The proposed diagnosis framework is validated via real experimental flight tests and has a reasonably high diagnosis accuracy.

Design and Experiment of a Deformable Bird-inspired UAV Perching Mechanism

Energy consumption and acoustic noise can be significantly reduced through perching in the sustained flights of small Unmanned Aerial Vehicles(UAVs).However,the existing flying perching robots lack good adaptability or loading capacity in unstructured environments.Aiming at solving these problems,a deformable UAV perching mechanism with strong adaptability and high loading capacity,which is inspired by the structure and movements of birds'feet,is presented in this paper.Three elastic toes,an inverted crank slider mechanism used to realize the opening and closing movements,and a gear mechanism used to deform between two configurations are included in this mechanism.With experiments on its performance towards different objects,Results show that it can perch on various objects reliably,and its payload is more than 15 times its weight.By integrating it with a quadcopter,it can perch on different types of targets in outdoor environments,such as tree branches,cables,eaves,and spherical lamps.In addition,the energy consumption of the UAV perching system when perching on objects can be reduced to 0.015 times that of hovering.