This article analyzes the evolution and current trends in aerial robotic manipulation, comprising helicopters, conventional underactuated multirotors, multidirectional thrust platforms equipped with a wide variety of manipulators capable physically interacting environment. It also covers cooperative manipulation interconnected actuated multibody designs. The review is completed developments teleoperation, perception, planning. Finally, new generation presented our vision future.

In this note, we consider the flocking of multiple agents which have significant inertias and evolve on a balanced information graph. Here, by flocking, mean that all move with common velocity while keeping certain desired internal group shape. We first show algorithms neglect agents' inertial effect can cause unstable behavior. To incorporate effect, use passive decomposition, decomposes closed-loop dynamics into two decoupled systems: shape system representing locked describing motion...

In this paper, we propose a novel framework, passive set-position modulation (PSPM), which enables us to connect (continuous-time) robot's position sequence of slowly updating/sparse (discrete-time) signal via the simple (yet frequently used in practice) spring coupling with damping injection, while enforcing passivity closed-loop robotic system. The PSPM modulates original such way that modulated is as close possible (i.e., maximum information recovery for better performance), yet only...

We propose a novel semiautonomous haptic teleoperation control architecture for multiple unmanned aerial vehicles (UAVs), consisting of three layers: 1) UAV layer, where each is abstracted by, and controlled to follow the trajectory of, its own kinematic Cartesian virtual point (VP); 2) VP which modulates VP's motion according commands local artificial potentials (for VP-VP/VP-obstacle collision avoidance VP-VP connectivity preservation); 3) through single remote human user can command all...

We propose a novel agreement framework for multiple (possibly heterogeneous) agents evolving on directed information graph with non-uniform delays. Our proposed can ensure of certain scalar quantity among the agents, as long 1) each agent, we design local control s.t. its closed-loop transfer function has unit gain at dc and strictly less than unity elsewhere; 2) globally reachable node (i.e. there exists path from it to every other nodes); 3) delays are finite constants. Rendezvous...

We propose a control framework for the bilateral teleoperation between single master robot and multiple cooperative slave robots with communication-delay in master-slave communication channel. Using passive decomposition, we first decompose dynamics of slaves into two decoupled systems while preserving energetic passivity: shape system describing grasping aspect, locked representing overall behavior slaves. Then, by locally controlling cancellation disturbances on it, secure tight (and...

We propose a novel aerial manipulation platform, an omnidirectional robot, that is capable of wrench generation with opportunistically distributed/aligned Sectional rotors. To circumvent the tight thrust margin and weight budget currently available rotor battery technologies, we design optimization framework, which maximizes minimum-guaranteed control force/torque for any attitude while incorporating such important useful aspects as interrotor aerointerference, anisotropic task requirement,...

Abstract Along with visual and tactile sensations, thermal sensation by temperature feeling on the skin can provide rich physical information environment objects. With a simple touch of objects, relative be sensed even objects differentiated different properties without any cue. Thus, artificially reproducing accurate/controllable haptic signals human epidermis will certainly major research area to reconstruct more realistic virtual reality (VR) environment. In this study, for first time,...

We show that the Lagrange dynamics of quadrotor-manipulator systems can be completely decoupled into: 1) center-of-mass in E(3), which, similar to standard quadrotor dynamics, is point-mass with under-actuation and gravity effect; 2) "internal rotational" quadrotor's rotation manipulator configuration, which assumes form robotic full-actuation no effect. Relying on this structure, we propose a novel backstepping-like end-effector tracking control law, allow us assign different roles for...

We propose a novel multi-rotor flying platform, ODAR (omni-directional aerial robot), which is fully-actuated (i.e., can assume arbitrary motion in SE(3) or generate control wrench se(3)) by six opportunistically distributed rotors, each driven reversible ESC (electronic speed controller) for bi-directional thrust generation. Due to this omni-directional generation, the system realize such powerful behaviors impossible with conventional platforms as: 360° photo/video shooting while holding...

We propose a novel robotic platform for aerial operation and manipulation, spherically connected multiquadrotor (SmQ) platform, which consists of rigid frame multiple quadrotors that are to the via passive spherical joints act as distributed rotating thrust generators collectively propel by adjusting their attitude force. Depending on number configuration, this SmQ can fully (or partially) overcome issues underactuation standard multirotor drones operation/manipulation (e.g., body-tilting...

A thermal haptic device was developed to replicate feelings for virtual reality applications.

We consider bilateral teleoperation of a wheeled mobile robot over communication channels with constant delays. Our main objective is to enable humans control the much as they drive car: i.e. by operating master haptic joystick, can linear velocity and heading angle robot, like do so gas pedal steering wheel. Passivity closed-loop system also enforced that, even delays, stably safely teleoperate force-reflection. A semi-experiment (i.e. real master/simulated slave) performed validate...

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> A detailed experimental comparison study of several published algorithms for motion and force control bilateral teleoperators, with emphasis on Internet-based teleoperation, is presented. The investigates the effects data losses, communication delays, environmental constraints a teleoperation system different techniques, which are based wave variables, Smith predictors, recent synchronization....

Predicting dam inflow is necessary for effective water management. This study created machine learning algorithms to predict the amount of into Soyang River Dam in South Korea, using weather and data 40 years. A total six were used, as follows: decision tree (DT), multilayer perceptron (MLP), random forest (RF), gradient boosting (GB), recurrent neural network–long short-term memory (RNN–LSTM), convolutional network–LSTM (CNN–LSTM). Among these models, model showed best results predicting...

Low-cost light scattering particulate matter (PM) sensors have been widely researched and deployed in order to overcome the limitations of low spatio-temporal resolution government-operated beta attenuation monitor (BAM). However, accuracy low-cost has questioned, thus impeding their wide adoption practice. To evaluate PM field, a multi-sensor platform developed co-located with BAM Dongjak-gu, Seoul, Korea from 15 January 2019 4 September 2019. In this paper, sample variation analyzed while...

In recent years, prompt tuning has proven effective in adapting pre-trained vision-language models to downstream tasks. These methods aim adapt the by introducing learnable prompts while keeping pretrained weights frozen. However, can affect internal representation within self-attention module, which may negatively impact performance variance and generalization, especially data-deficient settings. To address these issues, we propose a novel approach, Read-only Prompt Optimization (RPO). RPO...

Forests serve as crucial carbon sinks, yet quantifying cycle processes within forested watersheds is challenging due to inherent complexities, including multiple pools and variability. Dissolved organic (DOC) transport from forests significantly impacts drinking water quality since it interacts with chlorine form disinfection byproducts. Although the Soil Water Assessment Tool-Carbon (SWAT-C) has been widely used understand fluxes at watershed scale, model primarily evaluated in non-forested...

We propose a passive bilateral teleoperation control law for pair of n-degree-of-freedom (DOF) nonlinear robotic systems. The ensures energetic passivity the closed-loop teleoperator with power scaling, coordinates motions master and slave robots, installs useful task-specific dynamics inertia motion guidance, obstacle avoidance. Consequently, behaves like common mechanical tool. A key innovation is decomposition, which decomposes 2n-DOF into two robot-like systems without violating...

We propose a novel haptic teleoperation control framework for multiple unmanned aerial vehicles (UAVs) over the Internet, consisting of three layers: 1) UAV layer, where each is abstracted by, and controlled to follow trajectory of, its own kinematic virtual point (VP); 2) VP which modulates VP's motion according commands local artificial potentials (for inter-VP/VP-obstacle collision avoidance inter-VP connectivity preservation); 3) through remote human user can command all (or some) VPs'...