Managing Adaptive Versatile environments
Introduction
The MavHome Project (Managing an Adaptive Versatile Home) is focused on conducting research in smart home technologies from the aspect of treating an environment as an intelligent agent [1]. We seek to develop and integrate components that will enable the intelligent environments of the future. The goals of these environments are to maximize the comfort of the inhabitants, minimize the consumption of resources, and maintain safety and security.
Work in intelligent environments is an important step in the forward progress of technology. As computing becomes more pervasive and people’s lives become busier, advances in intelligent environments can aid by automating the simple things (e.g., lighting and HVAC control), work to actively conserve resources (reducing cost), and improve safety and security. Environments that sense their own well-being and can request repair or notify inhabitants of emergencies can save property and lives. Homes that can increase their own self-sufficiency over time can augment busy or aging inhabitants allowing people to live in their homes longer (potentially alleviating some health care system burdens) and free time to allow people to focus on other aspects of their lives. These are just some of the potential benefits of working intelligent environments, research and advancements in this area of science stand to make a large impact on the future.
The goal of this paper is to present one possible engineered approach to developing intelligent environments. We present the MavHome architecture, some of the lessons learned, some of our experimental results, and background work in this area of research.
Section snippets
Approach
Our work focuses on learning to automate the intelligent environment. The motivation for this work is the development of systems to meet this focus in an accurate and efficient manner. There are a number of significant challenges in this work in order to meet our goals, which are to learn a model of the inhabitant of an intelligent environment, automate devices to the fullest extent possible using this model in order to maximize the comfort of the inhabitant while maintaining safety and
Architecture
A decision-maker is our core control policy component. Our approach is to utilize an overall control algorithm in a three-phase system. The first phase will extract the appropriate observation data from a database and control the data-mining algorithm in order to find patterns and patterns of patterns to build a hierarchical hidden Markov model. This HHMM will be extended with actions and rewards to form a HPOMDP model of the inhabitant for the environment under evaluation. The observation data
Experimental environments
This work uses two real environments and their simulated counterparts. The MavPad is an on-campus apartment and the MavLab is the workplace of the researchers of this project. ResiSim is an in-house developed “residential simulator” for interactive simulation of intelligent environments.
Experimentation and results
We have implemented and tested our systems in the previously described environments and present some of our findings starting with simulation work with MavLab in ResiSim followed by MavPad experiments involving real inhabitants.
Related work
Our work focuses on the emerging domain of intelligent environments or smart homes and buildings. Generally, these environments are defined by the way in which people interact with them or in the way that these places interact with the inhabitants. Benefits include providing comfort and productivity for inhabitants and generating cost savings for utility consumption. There are many researchers working on interesting problems in this domain. Due to space considerations we provide a brief mention
Conclusions
Overall, our approach, design, and experimentation provide a level of environmental automation for both virtual and real inhabitants from a data-driven automatically learned model that can adapt to user pattern changes over time. The key strength of our work is that the model does not require a human to create the model or for knowledge to be created in the system for the model to be generated. A minimal amount of knowledge is required to automate and adapt—namely the automatable actions. Our
Future work
We continue to improve our approach and experiment with inhabitants in our environments. Our immediate goals are focused on techniques to assist with resource consumption reduction and to handle sensor noise. We are also working more in concert with the inhabitant to alleviate the control struggle and provide a more natural partnership between the inhabitant and the environment. We see our approach becoming a part of a system that works more closely with the inhabitant instead of just
Acknowledgement
This work was supported by National Science Foundation grants IIS-0121297 and EIA-9820440.
G. Michael Youngblood, a native of Bishop, Texas, earned his Ph.D., M.S., and Honors B.S. in Computer Science and Engineering from The University of Texas at Arlington. He is a member of Sigma Xi, Tau Beta Pi, Upsilon Pi Epsilon, and Omicron Delta Kappa. From 1988 to 1996, he served in the United States Navy Submarine Force. Dr. Youngblood is currently the Chief Scientist for the MavHome project and manages the AI Lab. His research interests include ambient intelligence, pervasive computing,
References (40)
- et al.
The role of prediction algorithms in the MavHome smart home architecture
Smart Homes
IEEE Wireless Communications
(2002) - L.P. Kaelbling, M.L. Littman, A.R. Cassandra, Planning and acting in partially observable stochastic domains, Tech....
- et al.
Artificial Intelligence: A Modern Approach
(1995) - et al.
The hierarchical hidden Markov model: Analysis and applications
Machine Learning
(1998) - G. Theocharous, K. Rohanimanesh, S. Mahadevan, Learning hierarchical partially observable Markov decision processes for...
- E.O. Heierman, Using information-theoretic principles to discover interesting episodes in a time-ordered input...
Zero configuration networking
- et al.
Reinforcement Learning: An Introduction
(1998) - E. Heierman, D.J. Cook, Improving home automation by discovering regularly occurring device usage patterns, in:...
Stochastic Complexity in Statistical Inquiry
(1989)
Circular Statistics in Biology
MIT Project AIRE
Ambient intelligence for the networked home environment
Interactive workspaces
CUED: Laboratory for communication engineering
[AHRI] — Aware Home Research Initiative
MIT Media Lab: Projects list database
Cited by (40)
Adaptive and context-aware privacy preservation exploiting user interactions in smart environments
2014, Pervasive and Mobile ComputingCitation Excerpt :However, these interactions, which also involve different sources of information, are usually very dynamic and complex. Hence, significant effort might be required to take autonomous decisions, as well as to adapt to individual users as they interact with the environment [8]. One of the foundations of pervasive systems and smart environments is the so called context, defined as the circumstances or background elements which (help to) determine the meaning of an event or the occurrence of specific patterns.
Smart home reasoning systems: a systematic literature review
2021, Journal of Ambient Intelligence and Humanized ComputingSensor-based activity recognition in the context of ambient assisted living systems: A review
2019, Journal of Ambient Intelligence and Smart EnvironmentsPersonalized adaptation in pervasive systems via non-functional requirements
2018, Journal of Ambient Intelligence and Humanized ComputingTrends and trajectories for explainable, accountable and intelligible systems: An HCI research agenda
2018, Conference on Human Factors in Computing Systems - ProceedingsSmart real-time meeting room
2017, TENSYMP 2017 - IEEE International Symposium on Technologies for Smart Cities
G. Michael Youngblood, a native of Bishop, Texas, earned his Ph.D., M.S., and Honors B.S. in Computer Science and Engineering from The University of Texas at Arlington. He is a member of Sigma Xi, Tau Beta Pi, Upsilon Pi Epsilon, and Omicron Delta Kappa. From 1988 to 1996, he served in the United States Navy Submarine Force. Dr. Youngblood is currently the Chief Scientist for the MavHome project and manages the AI Lab. His research interests include ambient intelligence, pervasive computing, entertainment computing, machine learning, and cognitive architectures.
Dr. Diane J. Cook is a professor in the Department of Computer Science and Engineering at the University of Texas at Arlington. She received her M.S. and Ph.D. in Computer Science at the University of Illinois, and conducts research in the areas of machine learning, data mining, intelligent environments, robotics, and parallel algorithms for artificial intelligence. She can be reached at [email protected].
Dr. Lawrence B. Holder is a professor in the Department of Computer Science and Engineering at the University of Texas at Arlington. He received his Ph.D. degree in Computer Science from the University of Illinois at Urbana-Champaign in 1991. His research interests include artificial intelligence, machine learning and data mining, in which he has had over 100 publications and support from DARPA, NASA and NSF. Dr. Holder is a member of AAAI, ACM and IEEE.