AI and ML at Heart of Hyundai's Autonomous Driving Venture

Share this article
Share this article
Prioritise Us on Google
Motional is a joint venture between Aptiv and Hyundai. Credit: Hyundai
Aptiv and Hyundai's venture has developed an end-to-end autonomous driving system that challenges conventional approaches to self-driving technology

Motional, Hyundai's autonomous driving venture, is preparing to launch a fully driverless Level 4 robotaxi service in Las Vegas – powered by an AI and machine learning system that represents a significant evolution in autonomous vehicle technology.

The joint venture between Aptiv and Hyundai has developed an end-to-end autonomous driving system that challenges conventional approaches to self-driving technology.

Traditional autonomous stacks separate perception, decision-making and control into multiple modules. Motional's system is transitioning toward machine learning-based end-to-end motion planning that integrates these steps into a single learned decision process.

Youtube Placeholder

This architecture enables the system to learn and output natural driving behaviours, which could move autonomous vehicles towards more adaptable driving capabilities. The approach blends safety compliance with AI-driven software evolution, ensuring that technology maturity is matched by operational readiness.

The technology has been through a phased validation programme which includes large-scale simulation of diverse scenarios, testing in closed environments and measured expansion on public roads.

Neural network training and validation

The aforementioned methodical approach reflects Motional's commitment to safety alongside technological advancement.

Laura Major, CEO of Motional, says: "Motional does not place safety behind technological progress. We remain focused on accelerating our ability to scale the technology and operations in a sustainable way, from development through the commercialisation stage."

Laura Major, President and Chief Executive Officer of Motional. Credit: Motional

The machine learning system must process data from more than 30 sensors, cameras, radars and lidar units integrated into the all-electric IONIQ 5 robotaxi. These sensors provide robust 360-degree perception, high-resolution images and ultra-long range detection of objects for safe autonomous operation in diverse driving environments.

Sensor fusion and perception

The robotaxi is a Society of Automotive Engineers (SAE) Level 4 autonomous vehicle that can operate safely without a driver. The vehicle is produced at Hyundai Motor Group Innovation Centre in Singapore and is built on the Group's Electric-Global Modular Platform (E-GMP), designed for ultra-fast charging and long-range driving.

According to Woongjun Jang, former Head of the Autonomous Driving Centre at Hyundai Motor Group and former Chief Strategy Officer at Motional, the integration represents a crucial milestone.

Woongjun says: "Hyundai Motor has evolved its IONIQ 5, a battery electric vehicle built on its EV-dedicated platform, into a platform for fully-autonomous vehicles. For the IONIQ 5-based robotaxi, we have applied various redundancy systems, in addition to a suite of essential technologies to ensure safety and convenience of the passengers.

"By successfully integrating the Group's IONIQ 5 robotaxi with Motional's autonomous driving technology, we are proud to announce that we have achieved another important milestone on the road to the commercialisation of our robotaxi."

Youtube Placeholder

The AI system must process inputs from the sensor array in real-time, making split-second decisions about navigation, obstacle avoidance and passenger safety.

The sensor fusion technology combines data from multiple sources to create a comprehensive understanding of the vehicle's environment, enabling safer and more reliable autonomous operation.

Commercial deployment with Uber

Beginning on 31 December 2026, the service, in collaboration with Uber, will allow users in Las Vegas to be matched with an all-electric Motional IONIQ 5 robotaxi. The partnership could provide valuable real-world data to further train and refine the machine learning models.

Sarfraz Maredia, President of Autonomous Mobility & Delivery at Uber, says: "We're excited to take the next step with Motional as we launch autonomous rides in Las Vegas.

Sarfraz Maredia, Head of Autonomous Mobility & Delivery at Uber. Credit: Uber

"This milestone reflects our shared commitment to introduce autonomous vehicles in a way that prioritises safety, increases reliability, and expands access to more ride options for our customers."

The Las Vegas deployment will serve as a proving ground for Motional's AI-driven approach to autonomous driving, testing whether end-to-end machine learning systems can deliver the safety and reliability required for commercial robotaxi operations.

Ultimately, the collaboration between Motional and Uber represents a significant step towards mainstream adoption of autonomous vehicle technology in urban environments.

Company portals

Executives