Skip to content
IFAI

Robotics

A driverless Maserati just broke the autonomous speed record

Published Reach 33

A fully driverless Maserati MC20 Coupe just hit 197.7 mph (318 km/h) at the Kennedy Space Center, shattering the autonomous land speed record. Powered by algorithms from Politecnico di Milano and Indy Autonomous Challenge, the team pushed the AI software to its absolute limit to stress-test systems and eventually improve high-speed highway safety.

A driverless Maserati just broke the autonomous speed record
Image: theverge.com

There is a new fastest robot in the world, and it looks like a sleek Italian supercar.

Last week at the Kennedy Space Center, a fully driverless Maserati MC20 Coupe hit 197.7 mph (318 km/h), officially setting a new autonomous land speed record. There was no human behind the wheel, no remote controller in hand—just software managing a twin-turbo V6 engine down a historic 2.8-mile runway.

The previous record stood at 192.2 mph, set back in April 2022 by an IAC AV-21 race car. The team behind both milestones is the Indy Autonomous Challenge alongside PoliMOVE, an engineering group from Italy's Politecnico di Milano.

Pushing street-legal metal to the edge

Unlike previous tests that relied exclusively on bespoke open-wheel race cars, this run used an actual production vehicle.

The Maserati MC20 retails for at least $239,000. It sports a mid-mounted 3.0-liter twin-turbo V6 engine dubbed the Nettuno, pushing 621 horsepower and 538 pound-feet of torque to the rear wheels through an eight-speed dual-clutch transmission.

Indy Autonomous Challenge CEO Paul Mitchell emphasized that this was far more than an expensive PR stunt. According to Mitchell, pushing self-driving software and robotics hardware to the absolute brink on a street-derived platform accelerates the transfer of racing innovations directly into highway-ready autonomous vehicles.

At 197.7 mph, latencies of just a few milliseconds can lead to catastrophic failure. Pushing the vehicle to its mechanical limits essentially acts as an extreme stress test for sensor fusion, predictive pathing, and high-frequency chassis control.

Why it matters

We think shifting from specialized formula racers to a production street chassis was a smart strategic move.

Purpose-built race cars operate in aerodynamic conditions that have very little to do with consumer vehicles. By proving that autonomous software can tame a 621-horsepower, rear-wheel-drive production coupe at nearly 200 mph, the team demonstrated remarkable real-time chassis dynamics control.

For everyday drivers, this does not mean your next robotaxi will be flying down the freeway at 200 mph. What it does mean is improved edge-case safety. If a vision and control stack can maintain stability under the violent lateral forces and aerodynamic instability of 318 km/h, it is vastly better equipped to catch a slide or react to sudden hydroplaning at normal highway speeds.

However, it is important to keep the setting in perspective. Blasting down an empty, perfectly flat NASA runway in a straight line avoids the hardest parts of self-driving: phantom braking, unpredictable pedestrians, and messy edge cases. High-speed stability is solved here, but urban autonomy remains the real battlefield.

Sources (1)

  1. [1]
    theverge.com(opens in a new tab)Originaltheverge.com

Best of AI, daily