Login | Register

Towards Smart Earthwork Sites Using Location-based Guidance and Multi-agent Systems


Towards Smart Earthwork Sites Using Location-based Guidance and Multi-agent Systems

Vahdatikhaki, Faridaddin (2015) Towards Smart Earthwork Sites Using Location-based Guidance and Multi-agent Systems. PhD thesis, Concordia University.

[thumbnail of Vahdatikhaki_PHD_F2015.pdf]
Text (application/pdf)
Vahdatikhaki_PHD_F2015.pdf - Accepted Version


The growing complexity and scope of construction projects is making the coordination and safety of earthwork of a great concern for project and site managers. The difficulty of safeguarding the construction workers is mainly commensurate with the type, scale, and location of the project. In construction operations, where heavy machines are used, various safety and risk issues put the timely completion of a project at stake. Additionally, the construction working environment is heavily susceptible to unforeseen changes and circumstances that could impact the project, both cost and schedule wise. As a response to the looming safety threats or unforeseen changes of working conditions, re-planning is almost always required, in both proactive (preemptive) or reactive (corrective) fashion. In order for re-planning to yield the optimum results, real-time information gathering and processing is a must. Global Positioning System (GPS) and other Real-time Location Systems (RTLSs) have been used for the purpose of real-time data gathering and decision-making in recent years. Similarly, Location-based Guidance Systems (LGSs), e.g., Automated Machine Control/Guidance (AMC/G), have been recently introduced and employed, mainly for the purpose of high-precision earthwork operations. However, currently the application of available LGSs (i.e., AMC/G) is restricted to the machine-level task control and improvement. Also, the high cost of procuring available LGSs, which cost approximately $80,000 for every new piece of equipment, limits the availability of LGSs for small and medium size contractors. Furthermore, the valuable real-time data gathered from various pieces of equipment on site are not effectively utilized to continuously update the simulation models developed at the design phase so that a more realistic view of project progress is available in the execution phase. Finally, despite the growing availability of LGSs, their application for safety is limited to real-time proximity-based object detection and warnings. In view of the ability to control the finest motion of LGS-enabled earthwork equipment, there is a great potential to boost their level of application to the project level, where decisions about the equipment control are made based on the global consideration of a fleet rather that a local view of one single equipment. To the best of the author’s knowledge, a generic methodology that combines real-time data-gathering technologies, LGS and intelligent decision making tools, particularly Multi-agent Systems (MASs), and addresses the safety-sensitive re-planning, is missing.
On this premise, this research pursues a methodology which addresses the issue of coordination and safety improvement through the integration of LGSs and MASs. In a nutshell, this research is dedicated to the pursuit of the following objectives: (1) to enable the project-level coordination, monitoring and control through the integration of a MAS architecture and a LGS to help better resolve operational and managerial conflicts; (2) to provide a method for improving the performance of pose estimation based on affordable RTLSs so that LGSs can be applied to a wider scope of older earthwork equipment; (3) to devise a generic framework for Near Real-Time Simulation (NRTS) based on data from LGSs; and (4) to develop a mechanism for improving the safety of earthwork operations using the capabilities of the LGS, NRTS, and MAS.
In the proposed framework, every staff member of the project is represented by an exclusive agent in the MASs. More affordable positioning technologies, such as Ultra-Wideband (UWB), are utilized to provide accurate real-time data about the location of machines and workers. An optimization-based method is proposed to consider a set of geometric and operational constraints that govern the behavior of the Data Collectors (DCs) attached to the equipment to improve the equipment pose estimation accuracy. NRTS is used to keep track of the progress of the project and fine-tune the schedule based on the data captured from the site. The agents observe the progress of work executed by their associated equipment, and if any anomalies are detected, viable corrective measures are devised and executed. The inputs to this system are: (a) a stream of real-time data, e.g., location data, flowing from the site, (b) the project design data, and (c) the project progress data and the schedule. Furthermore, a two-layer safety mechanism monitors the safe operation of different pieces of equipment. The first layer of this mechanism enables the equipment to plan a collision-free path considering the predicted movement of all other pieces of equipment. The second layer is acting as a last line of defense in view of possible discrepancies between the predicted paths and actual paths undertaken by the operators. Several prototypes and case studies are developed to demonstrate and verify the feasibility of the proposed framework.
It is found that the proposed optimization-based method has a very strong potential to improve the pose estimation using redundancy of more affordable RTLS DCs. Also, the proposed overarching NRTS approach provides a tracking-technology-independent method for processing, analyzing, filtering and visualizing the equipment states that can work with various types of RTLS technologies and under the availability of different levels of sensory data. The proposed safety system is found to provide a balance between economic use of space and the ability to warn against potential collisions in an effective manner using the pose, state, geometry, and speed characteristics of the equipment. Additionally, the safety system demonstrates the ability to provide a reliable basis for the generation of the risk maps of earthwork equipment, using the expected pose and state, and considering the proximity-based and visibility-based risks.
The MAS-based framework helps expand the effective domain of LGSs from machine-level guidance to fleet-level coordination. In the view of the presented case studies, the MAS structure is found to be effective in assigning different operations and tasks of a project to the specific agents that will be responsible for their realization. Using a combination of strategic and tactical planning methods, the MAS is able to effectively provide readily executable guidance/control for equipment operators considering a variety of safety issues.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Building, Civil and Environmental Engineering
Item Type:Thesis (PhD)
Authors:Vahdatikhaki, Faridaddin
Institution:Concordia University
Degree Name:Ph. D.
Program:Building Engineering
Date:20 July 2015
Thesis Supervisor(s):Hammad, Amin
ID Code:980205
Deposited On:27 Oct 2015 19:30
Last Modified:18 Jan 2018 17:50
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top