Risk-sensitive plan execution for connected sustainable home

Authors:
Masahiro Ono;Wesley Graybill;Brian C. Williams
Affiliations:
Keio University;-;Massachusetts Institute of Technology
Venue:
BuildSys '12 Proceedings of the Fourth ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Buildings
Year:
2012

Citing 6
Cited 1

Temporal constraint networks

Artificial Intelligence - Special issue on knowledge representation
Planning for temporally extended goals

Annals of Mathematics and Artificial Intelligence
TALplanner: A temporal logic based forward chaining planner

Annals of Mathematics and Artificial Intelligence
Coordinating agile systems through the model-based execution of temporal plans

AAAI'05 Proceedings of the 20th national conference on Artificial intelligence - Volume 1
An efficient motion planning algorithm for stochastic dynamic systems with constraints on probability of failure

AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 3
Survey Constrained model predictive control: Stability and optimality

Automatica (Journal of IFAC)

Probabilistic planning for continuous dynamic systems under bounded risk

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a risk-sensitive model-based plan executive, p-Sulu, as well as its application to energy-efficient room temperature control of a smart residential building called Connected Sustainable Home. p-Sulu minimizes the use of heating and cooling by optimally controlling the incoming sunlight through a south-facing facade made of electrochromic glass windows, whose opacity can be changed. It also helps the residents to choose energy-efficient behaviors by making suggestions on their schedule, such as when to leave home and when to go to bed. Furthermore, in the presence of uncertainties in the solar radiation and the outdoor temperature, p-Sulu robustly operates within a user-specified upper bound on the probability of constraint violations. The resulting robust optimal scheduling and control problem is solved by iterative solution of mixed-integer linear programmings. We demonstrate by simulations that p-Sulu significantly reduces the energy consumption compared to existing control approaches, while successfully limiting the risk of failure to satisfy residents' requirements.