Alternative Navigation

Navigating in areas where GNSS is degraded or unavailable usually requires expensive, specialised equipment. The challenge for the Plextek team was to create an accurate and low-cost alternative navigation system to enable a person to navigate accurately in the absence of GNSS.

The result was a system that attaches to a user’s boot and provides a highly accurate and robust position estimate when GNSS is ineffective or unavailable – inside buildings or tunnels, for example, or in an active GPS-denied environment.

The underpinning hardware and navigational algorithms offer low-drift alternative navigation capabilities in the event of GNSS signal loss.

Autonomous Drones are increasing being used for a wide variety of applications but are generally heavily reliant on GNSS for navigation. Our challenge was to develop an Alt-Nav sensor suitable for drones that could provide a GNSS-free navigation capability in all-weather day and night, and suitably small and low power.

In response Plextek developed a proof-of-concept terrain referenced navigation (TRN) system using our miniature PLX-T60 radar. It compares the elevation profile of the ground beneath the drone to a digital elevation map to estimate the Drone’s location when GNSS is degraded or unavailable.

Use of radar is advantageous as, unlike camera technology, it can work equally well day and night and is not affected by inclement weather such as rain or fog.

The challenge from the client was to take image retrieval to the next level – improving the performance of traditional visual scene matching (VSM) algorithms at finding the best matching image in a dataset of georeferenced images.

The Plextek team decided to augment the traditional VSM algorithms with one of the latest ML techniques – Semantic segmentation. This was used to assign a label to each pixel in an image, based on which category it belonged to – tree, person, vehicle, etc. This enabled the VSM algorithm to prioritise those features which are likely to remain consistent over time (such as buildings) over those that are likely to change (e.g. cars, vegetation). The result is a VSM algorithm that can more accurately estimate a system’s location by comparing imagery from a camera with a database of previously captured images.