LaneFree

Ce qu'il fait

LaneFree prevents deadly blind spot accidents in heavy vehicles through a retrofit camera system with AI detection that alerts drivers to hazards in real-time, addressing the critical safety gap in aging commercial fleets while generating safety analytics.

Votre source d'inspiration

The alarming statistics drove our solution: heavy vehicles cause 16.3% of accidents despite being only 6.5% of Singapore's vehicles, with fatal crashes rising 6% since 2023. Studying these incidents revealed blind spot collisions during lane changes were consistently deadly yet preventable. Existing solutions required expensive new vehicles or complex installations. We identified an opportunity to create an affordable retrofit system for Singapore's 91,000 heavy vehicles. The technology existed - it just needed adaptation for older fleets in a practical, accessible format that could be rapidly deployed to save lives immediately.

Comment ça marche

LaneFree is a modular, retrofit safety system built around two primary components: smart cameras and an intelligent processing unit. The system mounts wide-angle wireless cameras on vehicle exteriors to monitor blind spots, connecting to a Raspberry Pi inside the cabin. This unit runs our YOLOv8, which can identify vehicles, cyclists, and pedestrians in the camera feed in real-time. The true innovation lies in our bounding box assessment technique. The algorithm tracks not just the presence of objects but their relative movement and proximity, calculating collision risk based on the bounding box assessment. When a potential hazard is detected, the system delivers alerts visually and through sound. Installation is remarkably straightforward - the cameras attach externally with weatherproof mounts, while the main unit connects to the vehicle's 5V power supply. No integration with vehicle systems is required, making it compatible with vehicles of any age or model.

Processus de conception

We identified blind spots as a critical safety gap and mapped blind zone dimensions across commercial trucks. We architected a system using Raspberry Pi 4B as our processing hub, supporting wireless cameras with wide viewing angles specifically positioned to cover blind spots. Our design prioritizes universal retrofit capability with non-invasive mounting solutions that work across vehicle types. For the software design, we've selected YOLOv8 as our detection foundation, customizing it for transportation-specific object recognition. Our unique bounding box assessment algorithm distinguishes between vehicles traveling at matching speed versus collision threats, reducing false alarms. We've created detailed hardware integration diagrams, UI mockups featuring graduated alert visuals, and cloud connectivity specifications for data analytics. Currently, we're finalizing component sourcing while coding the core detection algorithms, preparing for our first functional prototype build.

En quoi est-il différent ?

LaneFree's uniqueness stems from three innovations: First, unlike factory-installed systems requiring costly fleet replacement, our universal retrofit design works with any existing vehicle regardless of age or make, with no-modification installation under 30 minutes. Second, our intelligent risk assessment goes beyond simple presence detection used by competitors. Our bounding box analysis evaluates object movement patterns and trajectory, distinguishing between parallel traffic (low risk) and intersecting paths (high risk). This reduces alert fatigue while providing graduated warnings scaled to actual collision probability. Third, we've created a comprehensive data ecosystem through cloud analytics, transforming individual units into a network of safety insights. We provide this at a breakthrough price point ($500/unit) that's 80% less than alternatives, making fleet-wide adoption economically viable across Southeast Asia's commercial vehicle market.

Plans pour l'avenir

Our immediate roadmap includes building our first functional prototype within three months, followed by driver testing and algorithm refinement. We'll pursue LTA certification while establishing partnerships with three major logistics companies for expanded field trials. Long-term, we'll integrate LiDAR sensors for enhanced performance in extreme weather and develop predictive risk modeling. We aim to create an interconnected safety ecosystem that reduces commercial vehicle accidents by 40% across Southeast Asia within five years, establishing LaneFree as the regional standard for retrofit vehicle safety.

Récompenses

1st Runner Up for IdeasJam 2025