Please use this identifier to cite or link to this item: https://rinacional.tecnm.mx/jspui/handle/TecNM/4481
Title: REINFORCEMENT LEARNING MODULE FOR PERSONALIZED ASSISTED PEDALING IN HUMAN-ELECTRIC HYBRID VEHICLES (PEDELEC)
Authors: GARCIA REYES, CARLOS RICARDO TONATIUH
Issue Date: 2021-09-10
Publisher: Tecnológico Nacional de México
metadata.dc.publisher.tecnm: Instituto Tecnológico de Chihuahua II
Description: This document contains the complete research for a Reinforcement (RL) Learning Artificial Intelligence (IA) for the automatic control of a PEDE LEC power assistance module. The RL IA will use trial and error methods to determine the best available assistance for the PEDELEC rider using as reference the Torque produced at the crankshaft and the rider’s heartbeat at one second intervals. This data w ill be transformed into an RL Environment where an agent will learn from no previous experience how to increase or decrease the assistance level to maintain the rider’s heartbeat at a comfortable level. The implementation of the RL will use the principles of TD(0) control and SARSA Prediction to achieve the goal.
metadata.dc.type: info:eu-repo/semantics/masterThesis
Appears in Collections:TESIS MAESTRIA EN SISTEMAS COMPUTACIONALES

Files in This Item:
File Description SizeFormat 
REINFORCEMENT LEARNING MODULE FOR PERSONALIZED ASSISTED PEDALING IN HUMAN-ELECTRIC HYBRID VEHICLES (PEDELEC).pdfTEXTO COMPLETO3.94 MBAdobe PDFView/Open


This item is protected by original copyright



This item is licensed under a Creative Commons License Creative Commons