Localization & Tracking in Constrained Environments
Wireless localization and tracking (L&T) has been extensively studied in the past. However, most of the existing works target applications that are fairly well resource-provisioned (infrastructure-support through anchors, powerful beacons, etc.). Taking L&T to the next level and making them available to practical applications like first responder (infrastructure-free) situations, or massive scale (low-cost, low-power beacons) inventory tracking, requires us to rethink these wireless solutions such that they can efficiently operate and provide high accuracies even under these resource-constrained environments. Indeed, there are multiple dimensions (cost, power, scale, etc. as shown in figure above) across which constraints may be faced in practical applications. In this thrust, we take a principled approach to the design of L&T solutions under various such challenging, resource-constrained regimes. In particular, we target (i) absolute tracking in dynamic indoor environments that are characterized by lack of static wireless anchors (i.e. infra-support) and where tracked entities are mobile -- this caters to first responder and future multi-player AR/VR gaming applications in everyday environments; and (ii) massive scale tracking, where thousands of entities need to be tracked accurately in real-time at an extremely low cost (cents) and power (micro-W) to make the application feasible -- this enables a new level of real-time supply-chain visibility for the era of warehouses and fulfillment centers.
TrackIO: Infrastructure-free Indoor Tracking of Mobile Entities
First responders, a critical lifeline of any society, often find themselves in precarious situations. The ability to track them real-time in un-known indoor environments, across multiple floors, significantly contributes to the success of their mission as well as their safety. In this project, we design and build TrackIO – a novel system that is capable of accurately localizing and tracking mobile responders real-time in large indoor environments spanning multiple floors. TrackIO leverages the mobile virtual infrastructure offered by unmanned aerial vehicles (UAVs), coupled with the balanced penetration-accuracy tradeoff offered by ultra-wideband (UWB), to accomplish this objective directly from outside, without relying on access to any indoor infrastructure. In a subsequent version, we have also designed a UAV-free version, that relies on a slightly different set of algorithms to make TrackIO deliver on its capabilities even in the absence of an UAV. This UAV-free version is currently undergoing successful trials with numerous first responder departments across the country. Below, you will find details on the UAV version, followed by the UAV-free version.
Towards a practical system, TrackIO incorporates four novel mechanisms in its design that address key challenges to enable tracking responders (i) who are mobile with potentially non-uniform velocities (e.g. during turns), (ii) deep indoors with challenged reachability, (iii) in real-time even for a large network, and (iv) with high accuracy even when impacted by UAV’s position error. TrackIO’s real-world performance reveals that it can track static nodes with a median accuracy of about 1–1.5m and mobile (even running) nodes with a median accuracy of 2–2.5m in large buildings in real-time.
A demo of TrackIO is available below:
"TrackIO: Tracking First Responders Inside Out", A. Dhekne, A. Chakraborty, K. Sundaresan, S. Rangarajan, USENIX International Symposium on Networked Systems Design and Implementation (NSDI), Feb 2019.
The UAV-free version of the solution relies on a different set of algorithms that allows TrackIO to first localize all the entities relative to each other in a cooperative manner; the topology embedding of which is then translated to absolute location estimates with the help of a single reference anchor. The reference anchor can be placed by the first responders on the entrance/exit door, as they enter the building, which then allows all the responders to be accurately tracked in real-time with respect to this reference. Such a UAV-free version is conducive for deployment with all first responder departments, including those without access to a UAV program. This version of the solution has been in active development, has gone through numerous successful trials with fire departments across the country and is closer to product commercialization. In future, one can expect such solutions on other verticals such as venue security, construction, etc.
Low-power Massive Scale Indoor Tracking
As e-commerce and the associated deployment of warehouses and fulfillment centers, grows at an unprecedented rate, the need for real-visibility into time supply chain has become more critical than even. Real-time management through tracking of thousands of products, assets and inventory forms a key component of this visibility equation. However, to enable accurate tracking on such a massive scale, we need to depart from conventional active beacon-based tracking technologies that are both cost-prohibitive as well as maintenance-prone (battery replacement). Our approach to this challenge is to design battery-free RF tags that are capable of harvesting energy from existing commodity devices in the environment, and generating high-bandwidth signals that can be used to localize and track to high accuracies of sub-meter. More details to come soon ...