Motion

"I got Motion"

Modivation

In commercial fleet management, fuel consumption is more than just another operational metric. According to the American Transportation Research Institute (ATRI), fuel consistently accounts for 20% to 28% of a motor carrier's total operating costs. Alongside narrow margins of just 3%-8%, a mere 2% difference in fuel efficiency across a large logistics fleet can lead to a 16% gain of net revenue. Standard routing algorithms typically optimize for time or distance, but heavily loaded vehicles interact with physics in ways that standard car-centric algorithms ignore.

What it does

Motion looks to tackle this problem by creating a heavy-duty routing API that optimizes for kinetic energy conservation rather than just the shortest distance or time. It calculates routes that minimize steep grades, reduce mandatory stops, and preserve vehicle momentum, directly translating physical efficiency into reduced fuel burn.

For developers, it functions as a drop-in replacement for the Google Maps API. By matching Google's API endpoints, parameter structures, and JSON responses, a logistics company can switch its existing dispatch dashboards over to Motion simply by updating the base URL. This instantly upgrades their fleet's routing logic to a physics-based model without requiring a frontend rewrite or breaking existing UI implementations.

How we built it

We built our platform as a heavy-vehicle-optimized routing engine and drop-in Google Maps replacement.

The backend runs on FastAPI, securely routed and exposed to our custom domain via Cloudflare. At the core of the system is a modified Valhalla instance. Instead of standard shortest-path algorithms, routing evaluates a custom physical-kinematic graph weight equation that incorporates vehicle mass, precise road grade, and momentum loss from stops.

To ensure high-fidelity environmental modeling, we aggregate spatial and topographic data from OpenStreet View and USGS elevation surveys.

Calculated graph weights and high-throughput routing requests are managed by a dual-Redis and PostgreSQL caching layer, which significantly reduces query latency for complex route generation.

Finally, the API layer acts as a translation proxy. It accurately emulates Google’s proprietary endpoint structures and JSON responses, allowing existing frontend interfaces to seamlessly render our optimized routes via a simple URL substitution without breaking existing implementations.

What's next for Motion

Our milestone is improving our data ingestion and filtering pipelines to support more dynamic route weights. We would like to integrate more robust traffic data and localized weather conditions for additional features like wet road friction and wind resistance.

We also plan to build out client-facing tooling. This includes a frontend UI dashboard that allows users to visualize route costs, view topographic data, and enter specific vehicle parameters.