Our Story

The Journey Behind US Patent 12,001,207

A Vision for the Future of Autonomous Navigation

When vision-based autonomous navigation technology development began in the mid-2010s, autonomous vehicles were largely a futuristic concept for most people. The patent (US 12,001,207, with a priority date of August 2019 and issued in June 2024) protects a camera-based safety system developed during research that explored various vision-based approaches. The autonomous vehicle landscape has evolved, with robotaxis and autonomous delivery vehicles entering deployment, supported by advancements such as Tesla's FSD v12.

Technical Approach: Vision-Based Navigation

The research explored vision-based approaches for autonomous vehicles, differing from what the developers observed as prevalent industry methods at the time, which emphasized lidar-driven, sensor-heavy systems with detailed mapping and object classification, though industry methods varied widely.

The research addressed the question:

How can an autonomous vehicle navigate, maintain its path, and avoid obstacles using vision as the primary input?

This question guided the research and development process. The resulting patent (US 12,001,207) specifically protects a dual-module safety system that compares live camera images with pre-recorded route images to determine navigation safety—a practical approach that differs from purely map-free navigation but reduces dependence on frequently-updated detailed mapping infrastructure.

Validation Through Industry Evolution

The broader industry shift toward camera-based approaches, including developments like Tesla's FSD v12 using camera-fed neural networks, demonstrates growing confidence in vision-first architectures. According to the developers' perspective, comparing this approach to challenges encountered by companies relying heavily on lidar and sensor-fusion approaches—such as Cruise's operational suspension in late 2023 following substantial investment—may suggest potential advantages of vision-first architectures, though direct causation is not established and multiple factors contribute to any company's operational challenges.

The Market Opportunity

Autonomy is primarily a software-based function, which can potentially offer higher margins compared to hardware, depending on business models and market conditions. An effective licensing model focuses on software providers through per-vehicle or subscription-based agreements.

Market Scale and Growth Potential

Based on various industry reports and estimates, which can vary and are subject to change, projections indicate growth in the autonomous vehicle market. Deployment examples, such as Tesla's FSD program involving hundreds of thousands of vehicles, show continued expansion. Industry projections vary depending on the definition of "autonomous vehicle," but suggest growth in vehicles with advanced autonomous capabilities in the coming decade.

Applications beyond passenger vehicles include light delivery vehicles in the autonomous delivery sector, commercial drones for inspection and delivery, and specialized robotics in humanoid and industrial contexts.

Camera-based approaches have demonstrated viability in certain contexts and implementations, with adoption occurring across various applications, though viability depends on specific use cases and regulatory approval.

Technical Achievement

Camera-Based Safety Architecture

The patented system (US 12,001,207) specifically covers:

• A dual-module architecture separating safety determination from vehicle control
• Live camera image comparison with pre-recorded route imagery
• Safety value calculation to determine when navigation instructions can be safely executed
• Visual navigation point recognition for high-level route following
• Safety-threshold mechanisms including human operator intervention requests

Prototype System Capabilities (demonstrated in research, some aspects covered by the patent):

• Real-time autonomous navigation using camera inputs as primary sensor
• Decision-making based on visual scene correspondence and uncertainty assessment
• Remote teleoperation and data collection for route training
• Route-specific learning from recorded visual data

Note: The research process explored various vision-based learning approaches. The patent specifically protects the dual-module safety system architecture as described above, rather than general end-to-end neural network training methods.

Technology Demonstration

The demonstration depicts an autonomous research vehicle navigating using vision-based controls, including path visualization and obstacle awareness.

Why This Patent Is Available for Licensing

The Regulatory Reality

The startup, which was self-funded, developed a business case in a rural area, obtaining customer commitments and local regulatory approvals. Challenges arose in obtaining final regulatory approval from the central government in the Netherlands. These challenges were due to the organization's limited size and resources.

The Scale Challenge

Deep learning AI involves data collection and computational resources, often more than anticipated. The approach has shown technical viability, but the investment for commercial scale exceeds the organization's capacity.

Key factors include:

• Regulatory processes often involve larger organizations
• AI systems require scaled data collection
• Neural network training needs computational infrastructure
• Generalization across environments involves real-world testing

Regulatory and scale requirements rendered independent development unfeasible.

What This Means for Licensees

These challenges may present opportunities for other organizations. Organizations with resources, scale, and regulatory experience may utilize the patented innovations.

Potential Applications

IP Protection: Incorporate vision-based navigation methods into patent portfolios

Competitive Positioning: Address potential infringement in camera-first systems

Defensive Coverage: Develop IP as the industry moves toward vision-based methods

Development Efficiency: Use existing IP instead of new development

Technical Approach Status

The patent is available for licensing and protects a camera-based safety determination system for autonomous navigation.

Parts of the industry appear to have shifted from skepticism about camera-only systems to validation of vision-first approaches, though debates on sensor fusion versus vision-only continue. Market developments suggest that autonomy can be achieved without exclusive reliance on lidar or sensor-heavy architectures in some demonstrated cases, such as certain deployments, though many systems still incorporate multiple sensors for redundancy. Different technical approaches exist, ranging from pre-recorded route matching (as in this patent) to generalized neural networks.

Technology Validation

These capabilities were validated through internal development and real-world testing according to developer testing; results may vary in broader applications and independent verification may be needed for commercial deployment.

Looking Forward: Industry Developments

The autonomous vehicle industry is undergoing changes. The industry debates sensor fusion versus camera-only approaches, and systems like Tesla's FSD v12 indicate that camera-based architectures may be viable for autonomous navigation.

Major trends observed in some companies, though hybrid approaches remain common:

• Shift from lidar-heavy to camera-first architectures (in certain implementations)
• Emphasis on software differentiation
• Expansion to delivery and robotics
• Regulatory developments for autonomous operations

Potential areas include:

• Licensing to automakers with autonomous technology
• Partnerships with AV software providers
• IP for camera-based systems
• Coverage for drone and robotics applications

Licensing Information

The patent encompasses research, development, and validation of a technical approach aligned with industry trends.

Elements Include

Technical Knowledge: Understanding of vision-based navigation

System Architecture: Architecture with operational experience

Patent Coverage: Coverage during industry shifts to camera-first methods

Licensing Structures: Various licensing structures available

About Licensing Opportunities

The focus is on suitable licensing arrangements. The aim is deployment of the technology at scale with appropriate value for the innovation.

Discussions may cover:

• Exclusive or non-exclusive arrangements
• Per-vehicle models
• Partnership opportunities

Applications

Autonomous vehicles affect transportation, goods movement, service delivery, and technology interaction. The patent serves as a component in this area.

The patent may apply to established automakers with autonomous capabilities, software companies developing AV systems, delivery companies with autonomous fleets, drone manufacturers, and robotics firms with mobile platforms, providing IP coverage for vision-based navigation systems.

Inquiries

Patent licensing inquiries involve technical and business discussions regarding licensing terms and patent applications.

Autonomous technology is under development. Discussions on the patent's potential applications are available.

For inquiries on licensing US Patent 12,001,207, contact information is provided on the contact page.