Problems We Solve Markets Radar Solutions Projects Why Navtech? Innovation Lab Support

AdvanceGuard and AI – How We Use Machine Learning Now and for the Future

Artificial Intelligence (AI) is a hot topic throughout the security sector, and various types of AI – such as ChatGPT, deepfake videos and facial recognition – are often in the headlines. But how does Navtech Radar use AI in its intelligent security solution?

When considering AI, it is important to be precise: the term ‘artificial intelligence’ covers a very broad technology that takes many forms. At Navtech we use a form of AI called machine learning (ML) in our security solutions. Machine learning uses data in ways that mimic human intelligence; specifically, it allows applications to improve their performance over time by ‘learning’ from inputs (such as radar data) in ways comparable to human learning from experience or reading.

Artificial Intelligence is very useful technology, but it brings challenges and limitations. At Navtech we are mindful of the privacy, data protection and system control issues involved: we do not allow AI to make decisions on behalf of our operators, for example. Also, some things can be done using ML, but are done equally well by traditional means. Machine learning takes up lots of computational power, so we sometimes use more traditional, but equally effective, approaches where these are more efficient.

Machine learning in Navtech Radar security software

We currently use ML in our security software, such as AdvanceGuard, for two key purposes:

  • To continually improve AdvanceGuard’s interpretation of complex data, including radar and camera data, and
  • To give the user ever-improving information, so that operators can make better decisions.

This use of ML leads to site-specific improvements for each AdvanceGuard installation. For example, as AdvanceGuard interprets radar data it becomes ever better at understanding what indicates a human intruder and what indicates something benign, such as a rabbit or deer. Consequently, AdvanceGuard doesn’t trigger when it sees a benign presence, which minimises nuisance alarms. However, when there is a genuine threat AdvanceGuard uses ML to process visual data, which gives the user visual verification and track intruders.

This continual improvement, allied with the fact that AdvanceGuard can be extensively customised by the user on site, makes it ever more effective at ensuring security.

Machine learning reduces nuisance alarms

Conducting field tests

Navtech keeps AI visible

Since AI is an evolutionary step in analytics, it has a long-term role in AdvanceGuard, which is an intelligent, rules-based system. Indeed, we are already working on higher-level ML solutions. However, the decisions that AdvanceGuard makes will always be transparent and under the control of the operator: we can ensure this because we only use AI/ML solutions we develop ourselves, specifically for AdvanceGuard. At Navtech we will only use AI to enhance AdvanceGuard’s effectiveness. When you use AdvanceGuard, you retain full control of its operation and of any security-related decisions made at all times.

Get in touch

If you are interested to know more about our radar solutions or are looking to replace existing systems, please share your details, and our team will be in touch shortly.

Get in touch