Real-time monitoring and validation of waste transportation using intelligent agents and pattern recognition
Files in This Item:
|Russell_ucd_5090D_10055.pdf||29.14 MB||Adobe PDF||Download|
|Title:||Real-time monitoring and validation of waste transportation using intelligent agents and pattern recognition||Authors:||Russell, Sean E.||Advisor:||Collier, Rem W
O'Hare, Gregory M.P.
|Permanent link:||http://hdl.handle.net/10197/6848||Date:||2015||Abstract:||Within Ireland and other Organisation for Economic Co-operation and Development countries there has been a growing problem of unauthorised waste activity. A report on this activity highlighted a number of problems. Of these, unauthorised collection and fly-tipping of waste is of particular concern due to the potential to cause pollution and health problems.This thesis presents the Waste Augmentation and Integrated Shipment Tracking (WAIST) system. WAIST utilises technologies from the area of pattern recognition, agent-oriented programming and wireless sensor networks to enable the monitoring and validation of waste transportation in near real-time.As components of the WAIST system, this thesis also introduces and evaluates two technologies. The first is the classification of object state based on accelerometer data, and the second is the use of agent-oriented programming languages as a high level abstraction for reducing ``programmer effort'' when implementing intelligent behaviours within WAIST.Both evaluations show positive results. In the classification component, an accuracy of 95.8% has been achieved in an eight class problem. In the agent component, students completed more tasks when using agents than when using Java. Additionally, subjective feedback highlighted a perception that problems were easier to solve using agents.Finally the WAIST system itself was evaluated over a number of simulated waste shipments based on a number of criteria. The results are very positive for the timeliness of the system, the ability to track stopping locations of the shipment, the accuracy when identifying illegal dumping and the efficient management of energy resources.||Type of material:||Doctoral Thesis||Publisher:||University College Dublin. School of Computer Science and Informatics||Qualification Name:||Ph.D.||Copyright (published version):||2015 the author||Keywords:||Intelligent agents; Intelligent transportation systems; Pattern recognition; Waste transportation; Wireless sensor networks||Other versions:||http://dissertations.umi.com/ucd:10055||Language:||en||Status of Item:||Peer reviewed|
|Appears in Collections:||Computer Science Theses|
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.