Kakao Mobility Targets Autonomous Driving, Robotics With Physical AI Engine

Kakao Mobility is leveraging physical artificial intelligence (AI) technology to expand beyond a simple commission-based platform into the future mobility business. The "field data from real roads" accumulated over the past decade through Kakao T, its integrated transportation service app, will serve as the foundation for advancing existing services such as taxi, designated driver, navigation, car rental and parking — and as core infrastructure for ushering in the era of autonomous driving and robotics. **Core Foundation of Physical AI: Accumulating 'Real-Road Field Data'** The biggest challenge in implementing physical AI is perfectly converting physical reality into digital information that AI can use as a basis for judgment. Urban environments, the main stage for the mobility industry, are fraught with countless unpredictable variables — weather changes such as snow and rain, jaywalking pedestrians, and illegally parked vehicles in narrow alleyways. For AI to respond as flexibly as humans in such unpredictable environments, learning from massive volumes of "field data" is essential. As Korea's No. 1 ride-hailing operator with more than 38 million registered users, Kakao Mobility has been accumulating such real-world mobility data on a daily basis. Vivid field data — including veteran taxi drivers' optimal route know-how, the complex late-night conditions on back roads experienced by designated drivers, and navigation route records — represents a unique asset that even global Big Tech companies cannot easily replicate. "We have not merely operated services but have accumulated the vast data on mobility demand, supply and traffic flow into digital data that AI can learn from," said Kim Min-sun, head of Kakao Mobility's autonomous driving business team. "The taxi and designated driver businesses we have operated are valuable services and, at the same time, a massive 'data pipeline' for perfecting autonomous driving intelligence." **Years of 'Spatial Intelligence Technology' Become the Eyes and Brain of Self-Driving Cars** In autonomous driving, the area Kakao Mobility has earmarked as its future growth engine, "ultra-precise positioning" and "spatial recognition" technologies are critical. For vehicles to navigate real roads without human intervention, systems that can perfectly replace human eyes and brains must be built. To this end, Kakao Mobility acquired "Kimgisa," a popular navigation app, in 2015 and transformed it into Kakao Navi, internalizing core map and navigation platform technology years ahead of global mobility companies such as Uber and Lyft, which relied on third-party map data. The company has also steadily advanced its big data-based "map matching" technology. This technology probabilistically corrects a vehicle's exact position, maintaining high coordinate accuracy even in urban "shadow zones" where satellite (GPS) signals are frequently lost — such as areas near high-rise buildings, tunnels and underground roads. In 2020, Kakao Mobility commercialized the world's first indoor positioning technology based on LTE (4G) signals and introduced a "multi-origin route finding" function that tracks the real-time positions of millions of taxis to predict which vehicle can arrive fastest. Industry analysts say these ultra-precise positioning technologies and route prediction algorithms will serve as a solid foundation for "core spatial intelligence" — enabling autonomous vehicles and delivery robots to find destinations without error and optimize routes in real time. "The spatial intelligence capability to pinpoint locations accurately even in complex environments will serve as a core sensory organ, helping autonomous vehicles and robots navigate urban areas seamlessly," said Lee Sung-woo, head of the spatial information planning team. **Securing 'Digital-Physical Hubs' for Autonomous and Robotic Ecosystems** For autonomous driving and robotic services to be commercialized, securing "spatial infrastructure" — places where devices can stay and be maintained — is also critical. Uber, the world's largest ride-sharing company, recently acquired North American parking reservation app SpotHero for the same reason. Kakao Mobility has established a strategy to utilize offline parking facilities as core hub infrastructure for future mobility. It has already built an indoor navigation environment at the COEX parking facility in Seoul and is operating Korea's first "robotic valet service" in Cheongju, North Chungcheong Province, in collaboration with HL Robotics. "In the future mobility era, parking spaces will go beyond simple parking lots to become critical infrastructure accommodating various physical AI-based transportation means, including autonomous vehicles and robots," said Kim Tae-sung, head of the P&C business division. The company is also accelerating the recruitment of talent in the autonomous driving field. It recently created a "Physical AI" tab on its recruitment page and is hiring experienced professionals with five or more years of experience across four autonomous driving roles — from hardware and electrical/electronic systems (HW/E&E), which form the backbone of autonomous driving systems, to software (AI/SLAM), which serves as the core brain. The company also plans to broadly recruit global specialists, including overseas-educated Koreans considering returning home. "Looking at the trajectory of the business portfolio Kakao Mobility has built over the past several years through the Kakao T platform, it becomes clear that its leap to becoming a physical AI company is a future that has been meticulously prepared over a long period," an industry official said.

Kakao Mobility Targets Autonomous Driving, Robotics With Physical AI Engine

Mine-Hunting Dolphins Return to Spotlight

Defense AI Will Determine Military Power

SK hynix Is Not the Benchmark for Bonus Standards

Gwanwoldang Return Marks Beginning, Not End

Parents Who Raise Top-Tier Children

Hyundai Motor Heads to Silicon Valley to Recruit Future Tech Talent

Smart Glasses Sales Hit 8.7 Million in 2024, Korea's Share Near Zero

SK Broadband to Train All Employees in AI, Target 400 "AI Performers"

Samsung Group ETF Surges 57% This Year on Chip, Battery Rally

Google, Marvell Launch Two AI Inference Chips; Pilot Production Next Year