Hyunsoo Yang

Dynamics and control of quadrotor with robotic manipulator

We show that the Lagrange dynamics of quadrotor-manipulator systems can be completely decoupled into: 1) the center-of-mass dynamics in E(3), which, similar to the standard quadrotor dynamics, is point-mass dynamics with under-actuation and gravity effect; and 2) the “internal rotational” dynamics of the quadrotors rotation and manipulator configuration, which assumes the form of standard Lagrange dynamics of robotic manipulator with full-actuation and no gravity effect. Relying on this structure, we propose a novel backstepping-like end-effector tracking control law, which can allow us to assign different roles for the center-of-mass control and for the internal rotational dynamics control according to task objectives. Simulations using a planar quadrotor with a 2-DOF arm are also performed to show the theory.

Hierarchical cooperative control framework of multiple quadrotor-manipulator systems

We propose an hierarchical control framework for multiple cooperative quadrotor-manipulator systems, which allows us to endow the common grasped object with a user-specified desired behavior (e.g., trajectory tracking, compliant interaction, etc.). To achieve this, our control framework consists of the following hierarchical layers: 1) object desired behavior design; 2) optimal cooperative force distribution; and 3) individual quadrotor-manipulator control based on object stiffness model, which can also take into account different dynamics characteristics of the (slower/coarse) quadrotor-platform and the (faster/fine) manipulator. Simulations of object transport and compliant interaction with three quadrotor-manipulator systems are performed to illustrate the theory.

Camera-GPS-IMU sensor fusion for autonomous flying

GPS-IMU based sensor fusion is widely used for autonomous flying, which yet suffers from the inaccuracy and drift of the GPS signal and also the failure with the loss of GPS (e.g., indoor flying). To circumvent this issue, in this paper, we propose a new framework for camera-GPS-IMU sensor fusion, which, by fusing monocular camera information with that from GPS and IMU, can improve the accuracy and robustness of the autonomous flying. For camera and GPS-IMU calibration, a new Kalman filter is also proposed, which runs in parallel with the state estimation EKF and also utilize multiple key frames generated from the camera information. An autonomous flying experiment is performed to validate the theory.

Haptic tele-driving of wheeled mobile robot over the internet via PSPM approach: theory and experiment

ABSTRACT We propose novel haptic tele-driving control frameworks of a wheeled mobile robot (WMR) over the imperfect Internet communication network with varying delay and packet loss. We consider both the dynamic and kinematic WMRs and their various tele-driving modes. By utilizing passive set-position modulation framework, we can guarantee two-port passivity or passivity/stability combination of the closed-loop tele-driving system with some theoretical performance measures. Experiments are performed to show the efficacy of the proposed frameworks using the Internet-emulated communication and a custom-built dynamic/kinematic WMR.

Multi-rotor drone tutorial: systems, mechanics, control and state estimation

We present a tutorial introduction to the multi-rotor unmanned aerial vehicles, often simply referred as drones. We first explain typical configuration, components and construction of the drones. We then provide basic kinematic and dynamic modeling of drones and their principle of flight. Some representative motion control techniques are then presented, which take into account the issue of under-actuation of the drones. State estimation problem of the drones, that is crucial for their proper flying, yet, should be done only by using onboard sensors and their sensor fusion, is explained. Some emerging research directions requiring capability beyond typical drones are also mentioned.

Preliminary results on vision-inertial state estimation for SmQT system

We present some preliminary results on state estimation of SmQT system, which consists of multiple UAVs. By using multiple set of IMU-camera and additional IMU, we estimate whole state of UAVs and tool. The estimator merge vision data and IMU measurements under mechanical constraint of SmQT system to increase tracking accuracy and robustness. Simulation is also performed to illustrate the results.

Fast ZMP and friction force calculation of mobile robot trajectory on uneven trajectory

In this paper, we develop a zero-moment-point and friction force calculation framework for wheeled mobile robots on uneven terrain to evaluate roll-over and slippage tendency of the given trajectory. The proposed framework using dynamic constraints and passive decomposition has advantages of low computations and implementation cost. Simulation results using CarSim is provided to analyze the performance of the proposed framework.

2-D cooperative localization with omni-directional mobile robots

In this paper, we propose cooperative localization scheme based feedback control for the multiple omni-directional mobile robots. Using the proposed cooperative localization scheme, localization capability of each mobile robot is enhanced, although some of the robots loss its absolute position measurement. Position error is bounded based on relative measurements which is unbounded for dead reckoning. Implementation results are also provided to validate the theory.

Preliminary results on quadrotor manipulation control

We propose controller for quadrotor with arm system. Quadrotor is underactuated which makes control problem difficult. Attached manipulator makes system dynamics complicate. To control end-effector position, we used passive decomposition and it provides decomposed dynamics. Using decomposed dynamics, we proposed backstepping-like controller to control end-effector position. Proposed controller achieve desired end-effector trajectory via underatuated system while moving manipulator. Simulation results are presented.