A Deep-Reinforcement-Learning Approach to the Peg-in-Hole Task with Goal Uncertainties
MetadataShow full item record
The thesis proposed a framework to train deep-reinforcement-learning agents for fine manipulation tasks with goal uncertainties. It consisted of three aspects: state-space formulation, artificial training-goals uncertainties, and progressive training. The framework was used in a simulation for two fine manipulation tasks, square Peg-in-Hole and round Peg-in-Hole. The resulting behaviours were then transferred a physical robotic manipulator and compared to traditional training methods. The deep-reinforcement-learning agents trained using the framework in this work outperformed those trained with definite goals.
Showing items related by title, author, creator and subject.
Interpreting Bounded Rationality in Business and Industrial Marketing Contexts: Executive Training Case StudiesWoodside, Arch; Lai, W.; Kim, K.; Jung, D. (2009)This article provides training exercises for executives into interpreting subroutine maps of executives’ thinking in processing business and industrial marketing problems and opportunities. This study builds on premises ...
Short, C.; James, E.; Vandelanotte, C.; Courneya, K.; Duncan, M.; Rebar, Amanda; Plotnikoff, R. (2014)© Springer-Verlag Berlin Heidelberg 2014. Purpose To explore demographic, health, social-cognitive and behavioural correlates of resistance training among post treatment breast cancer survivors. Methods A sample of 330 ...
Participation outcomes in a randomized trial of 2 models of upper-limb rehabilitation for children with congenital hemiplegiaSakzewski, L.; Ziviani, J.; Abbott, D.; MacDonell, R.; Jackson, G.; Boyd, Roslyn (2011)Objective: To determine if constraint-induced movement therapy (CIMT) is more effective than bimanual training to improve occupational performance and participation in children with congenital hemiplegia. Design: Single-blind ...