Electrical and Computer Engineering ETDs

Publication Date

7-1-2016

Abstract

By taking advantage of complementary communication technologies, distinct sensing functionalities and varied motion dynamics present in a heterogeneous multi-robotic network, it is possible to accomplish a main mission objective by assigning specialized sub-tasks to specific members of a robotic team. An adequate selection of the team members and an effective coordination are some of the challenges to fully exploit the unique capabilities that these types of systems can offer. Motivated by real world applications, we focus on a multi-robotic network consisting off aerial and ground agents which has the potential to provide critical support to humans in complex settings. For instance, aerial robotic relays are capable of transporting small ground mobile sensors to expand the communication range and the situational awareness of first responders in hazardous environments. In the first part of this dissertation, we extend work on manipulation of cable-suspended loads using aerial robots by solving the problem of lifting the cable-suspended load from the ground before proceeding to transport it. Since the suspended load-quadrotor system experiences switching conditions during this critical maneuver, we define a hybrid system and show that it is differentially-flat. This property facilitates the design of a nonlinear controller which tracks a waypoint-based trajectory associated with the discrete states of the hybrid system. In addition, we address the case of unknown payload mass by combining a least-squares estimation method with the designed controller. Second, we focus on the coordination of a heterogeneous team formed by a group of ground mobile sensors and a flying communication router which is deployed to sense areas of interest in a cluttered environment. Using potential field methods, we propose a controller for the coordinated mobility of the team to guarantee inter-robot and obstacle collision avoidance as well as connectivity maintenance among the ground agents while the main goal of sensing is carried out. For the case of the aerial communications relays, we combine antenna diversity with reinforcement learning to dynamically re-locate these relays so that the received signal strength is maintained above a desired threshold. Motivated by the recent interest of combining radio frequency and optical wireless communications, we envision the implementation of an optical link between micro-scale aerial and ground robots. This type of link requires maintaining a sufficient relative transmitter-receiver position for reliable communications. In the third part of this thesis, we tackle this problem. Based on the link model, we define a connectivity cone where a minimum transmission rate is guaranteed. For example, the aerial robot has to track the ground vehicle to stay inside this cone. The control must be robust to noisy measurements. Thus, we use particle filters to obtain a better estimation of the receiver position and we design a control algorithm for the flying robot to enhance the transmission rate. Also, we consider the problem of pairing a ground sensor with an aerial vehicle, both equipped with a hybrid radio-frequency/optical wireless communication system. A challenge is positioning the flying robot within optical range when the sensor location is unknown. Thus, we take advantage of the hybrid communication scheme by developing a control strategy that uses the radio signal to guide the aerial platform to the ground sensor. Once the optical-based signal strength has achieved a certain threshold, the robot hovers within optical range. Finally, we investigate the problem of building an alliance of agents with different skills in order to satisfy the requirements imposed by a given task. We find this alliance, known also as a coalition, by using a bipartite graph in which edges represent the relation between agent capabilities and required resources for task execution. Using this graph, we build a coalition whose total capability resources can satisfy the task resource requirements. Also, we study the heterogeneity of the formed coalition to analyze how it is affected for instance by the amount of capability resources present in the agents.

Keywords

Heterogeneous Robotic Agents, Robotic Networks, Coordinated Control, Connectivity Maintenance, Lift Maneuver

Document Type

Dissertation

Language

English

Degree Name

Electrical Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Committee Member (Chair)

Oishi, Meeko

Second Committee Member

Tapia, Lydia

Third Committee Member

Sadler, Brian

Download

Share

COinS