TY - JOUR
T1 - Survey of Model-Based Reinforcement Learning
T2 - Applications on Robotics
AU - Polydoros, Athanasios S.
AU - Nalpantidis, Lazaros
PY - 2017/1/26
Y1 - 2017/1/26
N2 - Reinforcement learning is an appealing approach for allowing robots to learn new tasks. Relevant literature reveals a plethora of methods, but at the same time makes clear the lack of implementations for dealing with real life challenges. Current expectations raise the demand for adaptable robots. We argue that, by employing model-based reinforcement learning, the—now limited—adaptability characteristics of robotic systems can be expanded. Also, model-based reinforcement learning exhibits advantages that makes it more applicable to real life use-cases compared to model-free methods. Thus, in this survey, model-based methods that have been applied in robotics are covered. We categorize them based on the derivation of an optimal policy, the definition of the returns function, the type of the transition model and the learned task. Finally, we discuss the applicability of model-based reinforcement learning approaches in new applications, taking into consideration the state of the art in both algorithms and hardware.
AB - Reinforcement learning is an appealing approach for allowing robots to learn new tasks. Relevant literature reveals a plethora of methods, but at the same time makes clear the lack of implementations for dealing with real life challenges. Current expectations raise the demand for adaptable robots. We argue that, by employing model-based reinforcement learning, the—now limited—adaptability characteristics of robotic systems can be expanded. Also, model-based reinforcement learning exhibits advantages that makes it more applicable to real life use-cases compared to model-free methods. Thus, in this survey, model-based methods that have been applied in robotics are covered. We categorize them based on the derivation of an optimal policy, the definition of the returns function, the type of the transition model and the learned task. Finally, we discuss the applicability of model-based reinforcement learning approaches in new applications, taking into consideration the state of the art in both algorithms and hardware.
KW - Intelligent robotics
KW - Machine learning
KW - Model-based reinforcement learning
KW - Policy search
KW - Reward functions
KW - Robot learning
KW - Transition models
UR - http://www.scopus.com/inward/record.url?scp=85010716228&partnerID=8YFLogxK
U2 - 10.1007/s10846-017-0468-y
DO - 10.1007/s10846-017-0468-y
M3 - Journal article
AN - SCOPUS:85010716228
SN - 0921-0296
VL - 86
SP - 153
EP - 173
JO - Journal of Intelligent and Robotic Systems
JF - Journal of Intelligent and Robotic Systems
IS - 2
ER -