Jump to content
  • Advertisement

R&D [SIGGRAPH 2018] Mode-Adaptive Neural Networks for Quadruped Motion Control

Recommended Posts


Awesome! Is the 3ms you show the required runtime for the ANN to generate motion from an input as seen on the buttons on top of the screen? / Do you think it can become faster?

Personally i think the result is worth the time, but game designers should start to think about how this tech can create new gaming experiences while dealing with limitations about quantity.

Share this post

Link to post
Share on other sites
Posted (edited)

No, this is the m/s (meters per second) as a user-control to specify how fast the character should move. The locomotion gait then automatically results from this ;)

Edited by Sebastian Starke

Share this post

Link to post
Share on other sites

Ouch, haha :)

But some performance numbers would be very interesting for game devs! (it's not mentioned int the paper either).

I work on physical simulation of bipeds. Here balancing is such a tight constraint that natural locomotion comes out automatically and no machine learning seems necessary for the basics. But for more complex motion beyond walking / running i think both approaches need to be combined.

Share this post

Link to post
Share on other sites

Yeah, these guys at UBC do some nice work - just look for DeepMimic, which will be presented at the same venue where we are presenting our quadruped research. However, the computational requirement for the network is about 2ms per frame on a gaming laptop. It runs on the CPU right now with the Eigen library - running it on the GPU gives a significant speedup of course. From what I have heard, NVIDIA made a demo using the PFNN on humans with ~100 characters in parallel - so I think it's already applicable for practical use :)

Share this post

Link to post
Share on other sites

It's really amazing to see how animation tools are evolving, can't wait to see what games can be made from this.

So what is the expected and hoped for path that this awesome software will take after SIGGRAPH? Are you planing on making your own software with this or is your aim for it to be used directly with game engines?

Looks brilliant.

Share this post

Link to post
Share on other sites

I'm just working on my PhD, that's all ;) However, I think there is a good chance that people might want to integrate this because it's directly developed inside Unity. We will continue researching on other relevant projects for character animation using AI techniques - so let's see... :)

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Advertisement
  • Advertisement
  • Popular Tags

  • Popular Now

  • Advertisement
  • Similar Content

    • By Dromo
      I am about to start a PhD that will investigate ways of replicating creativity in the AI systems of simulated people in virtual environments. I will research which psychology theories and models to use in order to achieve this, with a focus on creative problem solving.
      The aim of this project is to create virtual characters and NPCs that can create new solutions to challenges, even if they have never encountered these before. This would mean that not every possible action or outcome would need to be coded for, so less development resources are required. Players would encounter virtual people that are not bound by rigid patterns of pre-scripted behaviour, increasing the replay value and lifespan of games, and the accuracy of simulations.
      I am looking for companies or organisations that would be interested in working with me on my PhD, and I think computer games companies might be the most likely. I am trying to think of ways in which this new AI system might benefit games companies, or improvements and new types of games that might be possible. I am on this forum to ask for your thoughts and suggestions please, so I can approach games companies with some examples. 
      Thank you for your time and interest.
    • By Gourav Mishra
      Please help in completing the code. I am unable to use defined function
      import sys
      import nltk
      import random
      from nltk.tokenize import word_tokenize,sent_tokenize
      GREETING_KEYWORDS = ("hello", "hi", "greetings", "sup", "what's up",)
      GREETING_RESPONSES = ["'sup bro", "hey", "*nods*", "hey you get my snap?"]
      User_input = input ("User said:")
      type (User_input)
      def check_for_greeting(sentence):
          """If any of the words in the user's input was a greeting, return a greeting response"""
          words = word_tokenize (sentence)
          if words in GREETING_KEYWORDS:
      check_for_greeting(sentence = User_input )
    • By Lightness1024
      Hello gamedev, I am currently evaluating the worthiness of jumping into RD work for an automatic impostor system in our engine. In the past I've witnessed tremendous performance increase from such a system into the engine of LumenRT (which has to cope with unoptimized user created content). We're a little bit in the same situation right now. Possibly large fields with way too much data (high poly etc..).
      So if the engine would support auto-impostor-ing of stuff that'd be cool. Though, to make it a bit more modern, I was thinking that we could extend the parallax validity of billboards by storing the depth too, and render them using parallax occlusion mapping.
      So the invalidation could come after the camera has moved to a more radical angle than for traditional impostors. These exist techniques with full volumetric billboards that I am aware of, but they need the gometry shader to generate slices, and cost heavy voxel storage. I need something very light on the bandwidth to cope with switch/PS4 limitations.
      Can you point me to modern research on well balanced imposter techniques sounding like this ? or any idea you have on the matter.
    • By xiaohan wen
      If you have CROWDFUNDED the development of your game, which of the following statements do you agree with?
      1. I went out of my way to try to launch my game by the estimated delivery date
      2. I made an effort to launch my game by the estimated delivery date
      3. I was not at all concerned about launching my game by the estimated delivery date
      Hi there! I am an academician doing research on both funding success and video game development success.
      For those who have CROWDFUNDED your game development, it would be extremely helpful if you could fill out a very short survey (click the Qualtrics link below) about your experiences.
      The survey would just take 5 minutes and I’ll be happy to share my findings of what leads to crowdfunding success and how it affects game development based on an examination of 350 Kickstarter projects on game development in return.
      This is an anonymous survey and your personal information will not be recorded.
      Thank you very much in advance!
    • By Scanmaster_k
      Just unless you missed it, Humble Bundle is currently selling a book bundle regarding AI and machine learning.
      The bundle includes one book for UE4 and a lot of general books.
  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!