TY - JOUR AU - Bielskis, A. A. AU - Guseinoviene, E. AU - Dzemydiene, D. AU - Drungilas, D. AU - Gricius, G. PY - 2012/05/04 Y2 - 2024/03/29 TI - Ambient Lighting Controller Based on Reinforcement Learning Components of Multi-Agents JF - Elektronika ir Elektrotechnika JA - ELEKTRON ELEKTROTECH VL - 121 IS - 5 SE - DO - 10.5755/j01.eee.121.5.1656 UR - https://eejournal.ktu.lt/index.php/elt/article/view/1656 SP - 79-84 AB - The paper presents a vision of sustainable eco-social laboratory, the <em>ESLab</em> which might be used to speed up the process of development of the recently proposed by authors of the Smart Eco-Social Apartment.  It is presented the multi-agent model of the ambient comfort measurement and environment control system to be used for the development of the <em>ESLab</em>.  The human Ambient Lighting Affect Reward index, the <em>ALAR</em> index is proposed at the first time used for development of the Reinforcement Learning Based Ambient Comfort Controller, the <em>RLBACC</em> for the <em>ESLab.</em> The <em>ALAR</em> index is dependent on human physiological parameters: the<em> </em>temperature, the <em>ECG</em>- electrocardiogram and the <em>EDA</em>-electro-dermal activity. The fuzzy logic is used to approximate the <em>ALAR</em> index function by defining two fuzzy inference systems: the <em>Arousal-Valence System</em>, and the <em>Ambient Lighting Affect Reward (ALAR) System. </em>The goal of the <em>RLBACC</em> is to find such the environmental state characteristics that create an optimal comfort for people affected by this environment. The Radial Basis Neural Network is used as the main component of the <em>RLBACC</em> to performing of two roles - the policy structure, known as the <em>Actor</em>, used to select actions, and the estimated value function, known as the <em>Critic</em> that criticizes the actions made by the <em>Actor</em>. The <em>Critic</em> in this paper was used as a value function approximation of the continuous learning tasks of the <em>RLBACC</em>. Ill. 9, bibl. 7 (in English; abstracts in English and Lithuanian).<p>DOI: <a href="http://dx.doi.org/10.5755/j01.eee.121.5.1656">http://dx.doi.org/10.5755/j01.eee.121.5.1656</a></p> ER -