Utilizing Motion Matching with Deep Reinforcement Learning for Target Location Tasks

Autor:	Lee, Jeongmin, Kwon, Taesoo, Shin, Hyunju, Lee, Yoonsang
Rok vydání:	2024
Předmět:	Computer Science - Graphics
Druh dokumentu:	Working Paper
Popis:	We present an approach using deep reinforcement learning (DRL) to directly generate motion matching queries for long-term tasks, particularly targeting the reaching of specific locations. By integrating motion matching and DRL, our method demonstrates the rapid learning of policies for target location tasks within minutes on a standard desktop, employing a simple reward design. Additionally, we propose a unique hit reward and obstacle curriculum scheme to enhance policy learning in environments with moving obstacles. Comment: Eurographics 2024 Short Papers
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2403.15902 Zobrazit plný text záznamu View this record from Arxiv