Autonomous vehicle manufacturers are hitting a wall that no amount of testing miles can overcome: regulators demand proof that self-driving systems make safe, logical decisions, but the AI powering these vehicles operates as a black box. A new collaboration between Ignite by Forvia Hella and Oxford Semantic Technologies (OST) is attempting to solve this explainability crisis, which has become the primary blocker preventing vehicles from advancing to higher levels of autonomy.

The partnership arrives as regulatory authorities in the UK open applications for operators to run autonomous taxis, buses, and private-hire cars. However, leading autonomous vehicle providers continue to struggle with a fundamental problem: while their vehicles are improving at driving performance, manufacturers lack robust ways to prove safety, compliance, and decision logic at scale. This gap has created a bottleneck preventing the industry from progressing from Level 2 autonomy, where drivers remain legally responsible, to Level 3 and Level 4, where manufacturers assume liability for vehicle decisions.

What's the Difference Between Machine Learning and Explainable AI?

Most autonomous vehicles today rely on machine learning, which finds patterns in massive datasets and produces statistical outputs. The problem is that even engineers cannot fully explain why the AI made a particular decision in a specific situation. Explainable AI, by contrast, uses carefully curated expert knowledge and logical reasoning to solve complex problems, making decisions traceable and understandable.

The Ignite and OST collaboration leverages knowledge-based AI to translate traffic laws, originally written for human interpretation, into machine-readable rule sets. This approach enables manufacturers to generate what engineers call "deterministic evidence," proving that vehicle behavior is safe and compliant with regulations. The technology uses OST's RDFox knowledge graph database to provide a reasoning layer to autonomous vehicle systems, improving decision-making in complex situations and bridging the gap between traffic laws and live autonomous decision-making.

"Hella Ignite.Drive applies knowledge-based AI by translating traffic laws, originally written for human interpretation, into machine-readable rule sets. This enables manufacturers to generate deterministic evidence that demonstrates safe and compliant vehicle behaviour for European type approval," said Felix Kortmann, Chief Technology Officer at Ignite by Forvia Hella.

Felix Kortmann, Chief Technology Officer at Ignite by Forvia Hella

How Does Explainable AI Help Autonomous Vehicles Get Approved?

• Regulatory Transparency: Knowledge-based AI collects and maps vehicle decisions, allowing regulators to see exactly why a vehicle acts in a certain way under different road conditions and circumstances.
• Faster Approval Timelines: By reducing the need for manual, market-by-market rule coding, the technology helps autonomous vehicle teams move faster toward approval-ready deployment across different regions.
• Safety Evidence: The system provides a "white box" of data that software engineers can use to understand how AI makes decisions, improve safety features, and generate hard evidence for regulators on compliance with traffic rules.

The real-world impact of this explainability gap has become increasingly visible. Recent incidents involving autonomous vehicles driving into flooded roads highlight how current AI systems struggle with complex decision-making in unexpected situations. Without the ability to explain why a vehicle made a particular choice, manufacturers cannot convince regulators that the system will behave safely in edge cases.

"Autonomous vehicles currently use AI to make a whole range of decisions on the road, but at the moment, manufacturers are struggling to show why or how these decisions are made. RDFox can help with this major barrier to progress," said Peter Crocker, CEO of Oxford Semantic Technologies.

Peter Crocker, CEO of Oxford Semantic Technologies

Ian Horrocks, an Oxford University professor and OST co-founder, emphasized the broader implications of this technology. "The AV case study is a great example of how knowledge-based AI can enhance data-driven systems. A key advantage of the technology is traceability, where decisions can be linked back to the rules and logic that produced them. In the automotive space, this visibility can revolutionise go-to-market strategies, improving the compliance and safety of AVs," he noted.

Why Does This Matter for the Autonomous Vehicle Industry?

The explainability problem represents one of the most significant regulatory hurdles facing the autonomous vehicle industry. Manufacturers have logged millions of testing miles and demonstrated impressive driving performance, yet they remain unable to move from Level 2 to Level 3 and Level 4 autonomy because regulators cannot verify that the AI is making decisions based on sound logic rather than statistical patterns that might fail in novel situations.

This collaboration demonstrates how knowledge-based AI, which uses expert knowledge combined with logical reasoning, can be applied to the autonomous vehicle sector in ways that machine learning alone cannot. By making AI decisions explainable and traceable, manufacturers can provide regulators with the evidence needed to approve higher levels of autonomy. The technology also has the potential to reduce development lead times by minimizing manual rule coding for different markets, allowing companies to scale their autonomous vehicle deployments more quickly across Europe and beyond.