This system relies on innovative, patent-pending techniques that combine data from built-in smartphone sensors to identify key delivery moments without requiring extra hardware, wireless networks, or cellular connections. These techniques draw on principles from physics to analyze sensor information and generate dependable records of a driver's path inside buildings.
The technology identifies specific occurrences automatically, including parking locations, building entries, elevator movements, floor changes, and the exact point where a package is left at a customer's door. Although the precise sensors involved are not fully detailed publicly, the setup likely incorporates common smartphone elements such as accelerometers to sense motion, gyroscopes to monitor direction, and possibly magnetometers or barometers to detect shifts in the environment, like rising in an elevator or moving between floors. This combination enables highly accurate tracking in areas where global positioning systems fail, offering real-time insights that assist in settling disagreements about lost orders by supplying clear evidence of the delivery route.
To delve deeper into the sensor fusion aspect, this method involves merging inputs from various sensors to produce a more reliable outcome than any single sensor could achieve alone. Algorithms play a central role here, often using approaches like Kalman filters, which predict a device's position based on prior data and then refine it with new sensor readings to minimize errors. For instance, an accelerometer might detect steps or vibrations, but it can accumulate noise over time; fusing it with gyroscope data for orientation helps correct drift, while barometric pressure changes can confirm vertical movements. Doorstep's proprietary algorithms build on these foundations, tailored specifically for delivery scenarios to ensure efficiency and accuracy in complex indoor settings.
In comparison, Mapbox Navigation, a popular tool for mapping and directions in mobile and vehicle applications, also uses sensor fusion but with a different emphasis. Mapbox typically blends global positioning system signals, inertial measurement units for motion and orientation, and even camera inputs for augmented reality features, helping with turn-by-turn guidance on roads. This makes it excellent for outdoor travel, where it can snap locations to street networks and handle challenges like tunnels or city skyscrapers by enhancing signals with visual lane detection. However, Doorstep stands out by concentrating on indoor spaces without relying on global positioning or external aids, making it better suited for gig workers navigating apartment complexes or office buildings during deliveries.
Setting up the system is simple, involving just a few key steps: initializing the software development kit, beginning the tracking session, noting important delivery events, and ending the session. It operates quietly in the background on the driver's phone, working with nearly all iOS and Android devices, including professional scanners like the Zebra TC-77, while keeping battery and memory use low to prevent disruptions during long shifts. A forthcoming addition is navigation assistance that guides drivers from their parking spot straight to the drop-off area.
For delivery workers in the gig economy, this means less stress overall. It streamlines the detection of fraudulent claims and speeds up resolutions, which could result in fewer unfair deductions for supposed missing packages and more streamlined paths by examining time spent at each stop. While privacy specifics are not broadly detailed in public materials, the approach prioritizes collecting only the essential data for tracking events, avoiding unnecessary personal information from drivers or customers. This kind of advancement could transform how gig work operates, providing tools that make daily tasks smoother and more equitable.
