Hardware Design of a Tethered Robot Deployment and Recovery System 

Get Complete Project Material File(s) Now! »

Chapter 2 Literature Review

The scope of this thesis covers a broad spectrum of engineering topics as they relate to deploying and supporting a semi-autonomous payload from an autonomous helicopter. Be-cause of this, a variety of technical literature was reviewed to understand the terminology and state of the art of the wide scope of subject matter herein. Some of this literature consists of prior Master’s Theses published by colleagues at the Virginia Tech Unmanned Systems Lab, as their research has laid the groundwork for this mission. Other literature covers the much studied problem of suspending a payload from a helicopter platform.

Tethered Ground Robot Operations

Operation of a ground robot tethered to a helicopter is a central area of research for the Virginia Tech Unamnned Systems Lab. The lab has a variety of unmanned helicopters tted with controllers for autonomous ight. Historically, the lab has involved a team of mechanical, electrical, and computer engineering graduate and undergraduate students to equip these platforms with a variety of sensors to accomplish di cult missions. This sec-tion will review the work published by previous graduate students which directly relates to the ground robot deployment mission. This work includes ground robot design, operator situational awareness, landing zone detection, and vision based payload tracking.

Ground Sampling Robot Mission

Rose’s thesis [1] contributes several cornerstone elements to this system. Primarily, he designs a teleoperated robot platform capable of supporting localization and sampling hardware. This robot platform is designed to be deployed from and remain attached to an autonomous helicopter. The platform, depicted in Figure 2.1, features a di erential track system for movement and steering. It has a large payload bay to be equipped with a robotic arm and sampling apparatus. The vehicle is controlled remotely from serial commands generated on the Operator Control Unit (OCU), which is to be linked through the helicopter, and passed to an onboard controller.
Additionally, Rose contributes path planning algorithms to determine e cient routes around obstacles. From stereo image data obtained onboard the helicopter, his algorithm lters ter-rain based on slope and gradient, and calculates e cient paths for an autonomous controller or human operator to navigate. The ability to navigate and accomplish mission objectives quickly and e ciently is critical due to the helicopter’s limited endurance.

Helicopter Slung Load Operations

The lift and mobility characteristics of the helicopter give it the unique ability to hoist, transport, and often precisely place payloads. The use of helicopters to convey slung payloads is a widely studied practice. It has many applications, both military and commercial. The dynamics of the helicopter-slung load system is a complex area of research.
Lucassen and Sterk [2] perform a dynamic stability analysis of a slung load on a hovering helicopter. Three dimensional equations of motion are derived, neglecting aerodynamic loading on the slung load and constraining the helicopter to a vertical plane. The slung load is modeled as a simple pendulum with a single attachment point. Experimental results agree with the simulated stability analysis.
After Lucassen and Sterk, much of the work in this eld studies complex models for speci c aircraft and speci c tether con gurations. Because the platform used by Virginia Tech has demonstrated the ability to maintain ight stability with a single tether slung load, this thesis will not cover complex ight dynamics. A more relevant area of research, however, has stemmed from these studies. This is the study of actively damping payload oscillations using robotic platforms.
Bisgaard conducts research in topics relevant to controlling a slung load from an unmanned helicopter. His goal is to support a mine detecting mission, where a tethered sensor is towed a short distance o the ground. A state estimator is developed for payload position with sensor fusion of vision-tracking and inertial data [3]. His algorithm uses an unscented kalman lter, which incorporates nonlinear models of the plant and sensors. A control structure is developed which utilizes feedback to control helicopter motion in a way that reduces payload pendulation [4]. His results are impressive, as pendulation is reduces signi cantly compared to the open loop system.
Agrawal, et al [5] develop a novel six-cable suspended robot. The robot, which served as a platform to hold payloads, controls all suspension cables for active damping of sway and accurate payload transportation. Combining the fast manipulation of the cable suspended robot with the cumbersome motion of the helicopter, they are able to achieve more precise control than either system individually [6] .
Rosen, et al [7] develop an active aerodynamic stabilization mechanism for the helicopter-slung load system. This stabilization approach consists of two vertical aerodynamically loaded surfaces for which the angle of incidence is varied by a controller. Angular rates of the helicopter and slung load as well as load acceleration are used in a feedback controller. The results show that this solution is capable of stabilizing the helicopter-slung load system for a wide range of airspeeds.

Payload Control from Cranes

One of the problems investigated in this thesis is the control of a suspended payload that be-gins to oscillate. Because this a problem which occurs in handling cranes, a review of control techniques within this eld is conducted. Bobasu, et al [8] develop nonlinear algorithms for adaptive control a of handling crane. The crane has a translational actuator which controls the position of a vertical actuator. The algorithm is simulated and the adaptive controller outperforms the simple nonlinear controller.
Yanai, et al [9] use an inverse dynamics calculation for feedback control of a crane.  The crane model has a two axis translational actuator which controls a trolley where a winch is mounted. The control method is used to control a payload to follow a path while minimizing error due to sway. The controller is shown to be e ective.
Abdel-Rahman and Nayfeh [10] investigate pendulation reduction in boom cranes using cable length manipulation. The application is a ship mounted boom crane which is excited by oceanic waves. Their strategy involves reeling and unreeling the payload at near-resonance conditions to alter the dynamics. They develop two dimensional and three dimensional models and show their control strategy to be e ective in reducing payload pendulation. With the selection of appropriate reeling and unreeling speeds, their control strategy is particularly e ective.
Bockstedte and Kreuzer [11] develop a control technique which dampens payload oscillations using only vertical actuation. The technique, called modal coupling control, emulates an elastic pendulum. Two models are introduced: a simple pendulum model, consisting of a single tether, and a ying crane model, with multiple tethers. Both models are e ective in reducing oscillations of the suspended payload.
The single tether model is adapted and tested later in this thesis. I nd that the controls strategy for emulating the elastic pendulum is dependant on a small angle approximation for pendulum frequency and the absense of aerodynamic damping. Because a real system is subject to aerodynamic damping and may be required to perform in cases of larger angles, this control strategy could be ine ective. An alternative to this control strategy is introduced in this thesis which does not depend on the small angle approximation. This strategy uses the same principles introduced by Bockstedte and Kreuzer, but is more robust to large pendulation angles.

1 Introduction 
1.1 Motivation
1.2 Overview of Work
2 Literature Review 
2.1 Tethered Ground Robot Operations
2.2 Helicopter Slung Load Operations
2.3 Payload Control from Cranes
3 Hardware Design of a Tethered Robot Deployment and Recovery System 
3.1 System Overview
3.2 Electromechanical Analysis
3.3 Electrical Design
3.4 Future Design Considerations
4 Control Architecture and Low-Level Control 
4.1 Control Architecture
4.2 Position Control
4.3 Speed Control
5 Tether Tension Control 
5.1 System Model and Controller
5.2 Experimental Results
6 Damping Control of Payload Oscillation 
6.1 Background Information
6.2 Variable Length Pendulum Dynamics
6.3 Damping by Autoparametric Resonance
6.4 Recommendations for Hardware Implementation
6.5 Conclusions
7 Conclusion and Recommendations 
GET THE COMPLETE PROJECT
Tethered Payload Control from an Autonomous Helicopter