FUSION SPARSE AND SHAPING REWARD FUNCTION IN SOFT ACTOR-CRITIC DEEP REINFORCEMENT LEARNING FOR MOBILE ROBOT NAVIGATION. Jurnal Teknologi (Sciences & Engineering), [S. l.], v. 86, n. 2, p. 37–49, 2024. DOI: 10.11113/jurnalteknologi.v86.20147. Disponível em: https://journals.utm.my/jurnalteknologi/article/view/20147.. Acesso em: 26 nov. 2024.