AI Now Finds the Safest Drop-Off Point for Every Passenger

Autonomous vehicles are getting better at driving. Nobody has fully solved what happens next — the moment a passenger steps out and needs to get from the car door to the actual destination.

For most people, that gap is a minor inconvenience. For a wheelchair user, a visually impaired person, or an older adult, it's the part that determines whether the whole journey works or doesn't.

MaaS-Bridge, built by South Korea-based LBS Tech, is an AI system designed specifically for that gap. It received the CES 2026 Best of Innovation Award in the Travel and Tourism category for addressing something most mobility platforms overlook entirely: whether the place a vehicle stops is actually accessible to the person getting out. It's a small but critical piece of autonomous vehicle last mile accessibility that existing navigation systems were never built to handle.

The Gap Nobody Was Solving

Ride-hailing apps and autonomous mobility platforms calculate routes based on one assumption — a non-disabled adult who can walk anywhere from any curb. Current navigation services generate routes for non-disabled adult users, overlooking the diverse needs of wheelchair users, visually impaired passengers, seniors, and families who each require specific pedestrian features like ramps, tactile paving, or curb modifications.

That assumption breaks down fast in a world where autonomous vehicles carry everyone. A self-driving taxi that stops in the most efficient traffic position isn't necessarily stopping near a ramp, a tactile path, or a safe crossing. For people with visual impairments, wheelchair users, and older adults, usable pedestrian environments vary even when the destination is the same. Two drop-off points one block apart can represent completely different realities for someone with limited mobility.

The last mile of any journey — from vehicle to front door — has always been the hardest part of accessible autonomous mobility to solve. MaaS-Bridge is the first system built specifically around it.

How MaaS-Bridge Works

MaaS-Bridge conducts multi-dimensional analysis of road traffic flow, pedestrian infrastructure near destinations, and individual user profiles to propose optimal Mobility Points — the safest, most accessible location for a specific passenger to be picked up or dropped off.

The system reads three things simultaneously: what the road looks like from a traffic standpoint, what the surrounding sidewalk infrastructure actually offers — ramps, tactile paving, curb cuts, crossing signals — and what the individual passenger actually needs based on their mobility profile. It then calculates the spot where all three align.

LBS Tech defines these locations as Mobility Points, enabling users to continue safely and accessibly to their destinations via suitable pedestrian pathways after leaving the vehicle. The vehicle doesn't just stop somewhere convenient. It stops somewhere that works for the specific person inside it.

That distinction is what separates MaaS-Bridge from standard navigation. Most systems optimize for traffic. MaaS-Bridge optimizes for the passenger — accounting for who they are, what they need underfoot, and whether the infrastructure at that exact spot can support their journey continuing safely.

Where It's Already Running

LBS Tech completed pilot programs in Birmingham, UK, in partnership with the West Midlands Combined Authority and Aston University, with AR-based route guidance receiving strong user feedback. Proof-of-concept projects launched in Ho Chi Minh City, New York, and Barcelona to validate localization across different pedestrian environments. Commercial services are already running in Seoul, Sejong, and Busan in South Korea.

Each city tests a different piece of the autonomous vehicle last mile accessibility problem. Birmingham validated the AR-guided pedestrian routing. New York tested the system against dense urban infrastructure. Ho Chi Minh City and Barcelona confirmed localization worked across entirely different sidewalk environments and urban layouts.

LBS Tech and its UK partners have also begun a four-year joint initiative to build a digital twin-based pedestrian environment monitoring system and establish standardization frameworks for integrating pedestrian data with autonomous driving technology.

That four-year program matters beyond LBS Tech specifically. Standardizing how pedestrian accessibility data connects to autonomous vehicle systems could shape how every future mobility platform handles drop-off and pickup decisions — turning what MaaS-Bridge does today into an expected baseline for the industry.

Siwan Lee, CEO of LBS Tech, described the company's direction clearly: "CES 2026 represents a pivotal opportunity to accelerate our global partnerships with cities, government agencies, and autonomous mobility providers. Through MaaS-Bridge, we aim to demonstrate the practical importance of pedestrian technologies in the final stage of autonomous driving."