Train your self-driving car AI in Grand Theft Auto V – what could possibly go wrong?

Great training ground (and don't worry about all that violence)


Developers building self-driving cars can now take their AI agents for a spin in the simulated open world of Grand Theft Auto V – via OpenAI’s machine-learning playground, Universe.

The open-source MIT-licensed code gluing GTA V to Universe is maintained by Craig Quiter, who works for Otto – the Uber-owned startup that delivered 51,744 cans of Budweiser over 193km (120 miles) using a self-driving truck.

The software comes with a trained driving agent; all developers need is a copy of the game to get cracking. After that, programmers can swap out the demo AI model with their own agents to test their code and neural networks. Universe and Quiter's integration code takes care of the fiddly interfacing with the game.

Video games new and old provide great training grounds for developing reinforcement learning agents, which learn through trial and error – or rather, trial and reward when things go right. OpenAI's Universe was released in December, and is a wedge of open-source middleware that connects game controls and video displays to machine-learning agents so they can be trained in the virtual arenas.

GTA V is preferred by various research groups because it gives developers a realistic, detailed world for researchers to test their algorithms and models.

San Andreas – a fictional state in the game – is almost one-fifth the size of Los Angeles, and includes 257 different vehicles, seven types of bicycles, and 14 weather patterns, allowing researchers to test self-driving cars in urban areas without having to hit the actual road.

“[Integrating with Universe] was a perfect fit because GTA V runs in Windows, and Universe allows running the AI separate from the game; in Linux, for example, where most AI work is done,” Quiter told The Register.

Youtube Video

The video above demonstrates an AI running around in GTA V via Universe. The screen is split into separate windows to allow developers to inspect what frames are given to the agent, view any diagnostics from the bot, and so.

The simulated environment provides heaps of labelled data allowing the agent to start recognizing cars, bicycles, road signs, and other objects. Plus, the code can be modified to import a Tesla Model S into the world or explore a custom-designed town, Quieter added.

Just in case anyone is worried about AI agents becoming murderous thugs by hanging around the mean streets of GTA V, all the violence has been stripped from the system, we're told. ®

Similar topics


Other stories you might like

  • Despite global uncertainty, $500m hit doesn't rattle Nvidia execs
    CEO acknowledges impact of war, pandemic but says fundamentals ‘are really good’

    Nvidia is expecting a $500 million hit to its global datacenter and consumer business in the second quarter due to COVID lockdowns in China and Russia's invasion of Ukraine. Despite those and other macroeconomic concerns, executives are still optimistic about future prospects.

    "The full impact and duration of the war in Ukraine and COVID lockdowns in China is difficult to predict. However, the impact of our technology and our market opportunities remain unchanged," said Jensen Huang, Nvidia's CEO and co-founder, during the company's first-quarter earnings call.

    Those two statements might sound a little contradictory, including to some investors, particularly following the stock selloff yesterday after concerns over Russia and China prompted Nvidia to issue lower-than-expected guidance for second-quarter revenue.

    Continue reading
  • Another AI supercomputer from HPE: Champollion lands in France
    That's the second in a week following similar system in Munich also aimed at researchers

    HPE is lifting the lid on a new AI supercomputer – the second this week – aimed at building and training larger machine learning models to underpin research.

    Based at HPE's Center of Excellence in Grenoble, France, the new supercomputer is to be named Champollion after the French scholar who made advances in deciphering Egyptian hieroglyphs in the 19th century. It was built in partnership with Nvidia using AMD-based Apollo computer nodes fitted with Nvidia's A100 GPUs.

    Champollion brings together HPC and purpose-built AI technologies to train machine learning models at scale and unlock results faster, HPE said. HPE already provides HPC and AI resources from its Grenoble facilities for customers, and the broader research community to access, and said it plans to provide access to Champollion for scientists and engineers globally to accelerate testing of their AI models and research.

    Continue reading
  • Workday nearly doubles losses as waves of deals pushed back
    Figures disappoint analysts as SaaSy HR and finance application vendor navigates economic uncertainty

    HR and finance application vendor Workday's CEO, Aneel Bhusri, confirmed deal wins expected for the three-month period ending April 30 were being pushed back until later in 2022.

    The SaaS company boss was speaking as Workday recorded an operating loss of $72.8 million in its first quarter [PDF] of fiscal '23, nearly double the $38.3 million loss recorded for the same period a year earlier. Workday also saw revenue increase to $1.43 billion in the period, up 22 percent year-on-year.

    However, the company increased its revenue guidance for the full financial year. It said revenues would be between $5.537 billion and $5.557 billion, an increase of 22 percent on earlier estimates.

    Continue reading

Biting the hand that feeds IT © 1998–2022