How LiDAR Boost supports GNSS-denied navigation

Getting accurate localisation data in areas where GNSS signal is weak or nonexistent is a challenge faced by a huge range of engineers and navigation professionals. At OxTS, we’ve just launched WayFinder, a brand-new product, alongside LiDAR boost, a brand new software tool designed to help solve this problem.

WayFinder is an all-in-one solution for GNSS-denied navigation that works out of the box, featuring integrated LiDAR and cameras in addition to a top-performing MEMS GNSS/INS.

In this blog, we’re getting into the detail of the unique software that powers Wayfinder: LiDAR Boost. Let’s start by recapping how people are currently trying to solve the GNSS-denied navigation challenge.

xxx

Sensor fusion is the future of navigation in challenging environments.

Navigating in areas where GNSS signal isn’t reliable or available isn’t a new challenge: organisations have been grappling with it for a number of years, as have we. In that time, a variety of solutions have been developed for navigating in indoor or underground environments, including using ultra-wideband (UWB), simultaneous localisation and mapping (SLAM), and solutions relying on magnetic strips in the ground.

These solutions all have their uses and are effective in many ways, but also come with challenges:

1. Some solutions rely on infrastructure being installed in the environment to work. This limits the operation of the vehicle to the area where the infrastructure is installed, which might not suit instances where testing doesn’t happen in a fixed space.
2. Many of these solutions often don’t perform as well in outdoor environments as they do indoors (with the exception of magnetic strips). Using these solutions still leaves you with the challenge of moving between environments where you can navigate with GNSS data, and environments where there is no GNSS signal.

At OxTS, we are championing the real-time fusion of additional sensor data into a GNSS/INS to solve the challenge. By feeding data from sensors such as LiDAR, using our LiDAR Boost software, into OxTS’ navigation engine in our GNSS/INS (where it can be combined with our inertial navigation data), we can provide accurate navigation data in environments such as warehouses without needing to install infrastructure.

Crucially, because the fusion is happening in the GNSS/INS rather than in a different piece of hardware, it’s possible to switch seamlessly to GNSS-aided navigation in areas where signal is strong enough. The result? A flexible navigation system that can handle any environment and move between environments with ease.

How does LiDAR Boost work?

Without GNSS signal, any inertial navigation system’s position data will start to drift as errors from the inertial measurement unit (IMU) begin to accumulate. LiDAR Boost constrains that drift to protect the accuracy of the data the GNSS/INS produces.

There are two main ways in which LiDAR Boost takes the data from a LiDAR scanner and turns it into data that an OxTS GNSS/INS can use:
xxx

– Protect accuracy when GNSS is patchy or inaccurate with real-time LiDAR Inertial Odometry (LIO)

LIO is an existing OxTS technology that analyses the data from a LiDAR scanner to detect planes and objects and track how they move between the ‘frames’ the LiDAR captures. That data is used to calculate relative velocity updates that an OxTS GNSS/INS can use. You can learn more about it in this blog.

LiDAR Boost takes that capability and executes it in real time. It’s ideal for areas where GNSS signal is patchy, for temporary outages like tunnels, or to protect against multipath errors that are common when navigating through cities.

 

– Achieve accurate and completely repeatable performance with LiDAR Map Matching (LMM)

If you have created a point cloud of a space, LMM can compare the data being gathered by the LiDAR scanner with that point cloud to produce position updates within that space. Because those position updates are based on the point cloud data, the results are completely repeatable – there is absolutely zero drift or variation.

LMM is therefore perfect for activities such as ground truthing and autonomous navigation, where being able to guarantee repeatable results is vital.