RT Conference Proceedings
T1 Reducing the number of evaluations required for CGDA execution through Particle Swarm Optimization methods
A1 Fernández Fernández, Raúl
A1 Estévez Fernández, David
A1 González Víctores, Juan Carlos
A1 Balaguer Bernaldo de Quirós, Carlos
AB Continuous Goal Directed Actions (CGDA) is a robot learning framework that encodes actions as time series of object and environment scalar features. As the execution of actions is not encoded explicitly, robot joint trajectories are computed through Evolutionary Algorithms (EA), which require a large number of evaluations. The consequence is that evaluations are performed in a simulated environment, and the optimal robot trajectory computed is then transferred to the actual robot. This paper focuses on reducing the number of evaluations required for computing an optimal robot joint trajectory. Particle Swarm Optimization (PSO) methods have been adapted to the CGDA framework to be studied and compared: naíve PSO, Adaptive Fuzzy Fitness Granulation PSO (AFFG-PSO), and Fitness Inheritance PSO (FI-PSO). Experiments have been performed for two representative use cases within CGDA: the “wax” and the “painting” action. The experimental results of PSO methods are compared with those obtained with the Steady State Tournament used in the original proposal of CGDA. Conclusions extracted from these results depict a reduction of the number of required evaluations, with simultaneous tradeoff regarding the degree of fulfillment of the objective given by the optimization cost function.
PB IEEE
SN 978-1-5090-6235-5
YR 2017
FD 2017-04-26
LK https://hdl.handle.net/10016/34067
UL https://hdl.handle.net/10016/34067
LA eng
NO Proceedings of: 2017 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), 26-28 April 2017, Coimbra, Portugal.
NO The research leading to these results has received funding from the RoboCity2030-III-CM project (Robtica aplicada a la mejora de la calidad de vida de los ciudadanos, fase Ill; S2013/MIT-2748), funded by Programas de Actividades I+D en la Comunidad de Madrid and cofunded by Structural Funds of the EU, and by a FPU grant funded by Miniesterio de Educaci6n, Cultura y deporte.
DS e-Archivo
RD 27 jul. 2024