NDC 메뉴로 바로가기 NDC 본문으로 바로가기

NDC - NEXON DEVELOPERS CONFERENCE

바로가기
LOGIN
KOREAN
ENGLISH
트위터
페이스북
인스타그램
How to train your robot - Massive Robots with Reinforcement Learning at Embark
At Embark Studios we have been building a system for Physically Based Animation of Large Virtual Robots. The goal of this is to create more emergent and interesting physical gameplay. Just like Physically Based Rendering improves fidelity and realism of graphics, Physically Based Animation improves the quality and realism of movements. On top of this, our players are also able to interact with these animations -- a heavy impact from a player will create a unique and believable response without any designer input. In this talk, we will describe the progress we have made so far, the possibilities and challenges we have seen, and what we hope to gain in the future.
  • 발표내용의 난이도 사전지식 불필요 : 관련 전공이나 경력이 전혀 없더라도 이해할 수 있는 내용
  • 대상 Game Designers , Animators , Technical Animators , Animation Programmers
  • 발표 시간 16분
  • 키워드 Physics, Emergent Gameplay, Robotics , Animation , Reinforcement Learning
Tom Solberg | 엠바크 스튜디오 / Embark Studios

발표자 소개

Software Engineer at Embark Studios, working on machine learning tools and infrastructure.

Top