It's safe to say that GTA V has been one of the most successful open world video games to date and thanks Rockstar Games' consistent updates, the game still thrived and even triumphs over other fresh new titles throughout the years. Today, it was revealed that not only does the game provides surreal experience to its fans, it now also serves as a training ground for self-driving cars which saves millions of dollars in return.
GTA V Saves Millions For Driving AI Simulations
GTA V offers more than your typical open world action-adventure game where you can drive tanks and cars and ram it to another one's garage or set them on fire for your delight since the game is now being used as cheap alternative for driving simulations which is thought to have saved millions of dollars in the process.
According to Universe Blog post, the self-driving car AI called DeepDrive Universe is now being tested on GTA V to measure its overall capacity, intelligence and proficiency on various situations. As shown on the video down below, the AI has integrated with game's driving system in which it provides real-time approach on a given scenario within the game. The intelligence of the AI is then measured depending on how accurate and how efficient it reacts on a given scene.
One of the developers of the program said: "Its island setting is almost one fifth the size of Los Angeles, giving access to a broad range of scenarios to test systems. Add to that the 257 different vehicles, 7 types of bicycles, and 14 weather types, and it's possible to explore a huge number of permutations using a single simulator."
It's interesting to know that GTA V, the game that's been so popular amongst all generations is also good for another reason aside from making cars explode across the streets. According to the post, the researchers have saved millions of dollars that would have otherwise been used to pay companies that make exclusive simulators for these AIs. Since the DeepDrive is now open source, you can now test the project yourself by following their blog post here.
