ORAYA — Relation Intelligence for Physical Systems

 

A new mathematical architecture for knowing when evidence is ready.

ORAYA — Origin Relational Authentic Yield Arising — determines not only what a sensor model predicts, but whether the available evidence has earned the right to support that endpoint, whether the system should abstain, and when the relation first becomes trustworthy.

​

Invented by Rocky Latchman

Developed at the PALA Institute for Spiral Cognition

U.S. Patent Pending — Provisional Application Filed

​

Externally Reviewer-Attested · Outcome-

 

Blind Result

 

ORAYA has progressed from mathematical formulation to implementation, public physical-data testing, blind whole-context evaluation and external-reviewer attestation.

In its strongest current test, ORAYA was required to generalise relation judgment to three wholly hidden physical boundary contexts without access to the hidden endpoint values, admission answers, first-admission answers, relation ratios, correlations or decoy scores.

23 / 23

Declared criteria passed

3 / 3

Whole-context hidden-fold wins

0.018519

False-admission rate

0.979058

Abstention accuracy

+0.055556

Risk-sensitive utility advantage over the fixed admission rule

0

Decoy false admissions

The evaluation covered 216 hidden context–start–endpoint relations across three completely held-out physical contexts.

​

What Is ORAYA?

Most artificial-intelligence systems are designed to produce an answer.

ORAYA asks a prior question:

Has the available evidence earned the right to support that answer?

A physical signal can contain information before that information is sufficiently reliable to justify a decision. A relation may be trustworthy under one operating condition and invalid under another. An endpoint may emerge early, emerge late, or never become admissible under the present conditions.

ORAYA is designed to identify those differences.

It evaluates the developing signal, the endpoint being considered and the surrounding physical context before determining the appropriate relation state.

Depending on the evidence, ORAYA distinguishes four possible relation states:

STATE 01 · ADMIT

The evidence is ready.

The available signal has reached sufficient support for the endpoint relation.

The relation may be used under the declared conditions.

STATE 02 · WAIT

The relation is developing.

The endpoint may become trustworthy as more evidence appears, but acting now would be premature.

ORAYA continues to observe the developing relation.

STATE 03 · ABSTAIN

The evidence does not justify the claim.

The scientifically appropriate result is to withhold the endpoint rather than force a prediction.

STATE 04 · NEVER IN THIS CONTEXT

The relation does not become admissible under the present conditions.

​

ORAYA can distinguish an endpoint that is merely early from one that should not be asserted within the observed context.

​

​

The Latchman Admission Field

At the mathematical foundation of ORAYA is the Latchman Admission Field, originated by Rocky Latchman.

Its public mathematical signature is:

LAFₗ : (Xₛ, Eⱼ, H) ⟼ (Aₗ, τₗ)

In plain language:

(Xₛ) represents the physical evidence available at a particular stage.

(Eⱼ) represents the endpoint under consideration.

(H) represents the permitted physical context.

(Aₗ) represents admission or abstention.

(τₗ) represents the first stage at which the endpoint becomes trustworthy — or the conclusion that no admissible onset exists in that context.

The proprietary mathematical construction beneath this signature forms part of ORAYA's protected intellectual property.

A New Applied Mathematics of Relation Trust

Traditional predictive mathematics asks:

What output follows from this input?

ORAYA introduces another mathematical problem:

Under what conditions has the relation itself become trustworthy?

That changes the object being studied.

ORAYA is concerned with the structure of:

evidence

→ relation validity

→ first trustworthy onset

→ admission or abstention

A physical relation does not have to be treated as permanently true or false.

It can emerge as a signal develops.

Its validity can change with physical context.

Its first trustworthy point can move.

And under some conditions, the relation may never become admissible.

This is why ORAYA's mathematics reaches directly into real-world decisions.

It asks not only what is likely, but when the evidence is sufficient to justify action.

​

Built for the Real World

ORAYA is designed as a relation-intelligence layer that can operate above or alongside existing predictive systems.

A company may already possess powerful machine-learning models and vast quantities of sensor data — yet still face a critical question:

When should this prediction actually be trusted?

ORAYA is being developed to address that gap.

Aerospace

Determine when developing vibration, structural or engine evidence becomes sufficiently trustworthy to justify an endpoint or intervention.

Predictive Maintenance

Separate genuine emerging degradation from premature or unsupported relations.

Identify when warning evidence first becomes actionable.

Non-Destructive Testing

Determine when a developing acoustic or guided-wave response begins to support a damage-sensitive endpoint.

Energy Systems

Evaluate when electrical, thermal, vibration or degradation signals become consequential rather than merely unusual.

Manufacturing

Identify which sensor relations remain valid as machines, materials, loads and operating conditions change.

Robotics

Determine when partial sensory evidence has become sufficiently trustworthy to support physical action.

Digital Twins

Add context-sensitive relation validation to simulations, monitoring systems and predictive models.

Scientific Instrumentation

Identify when measured evidence has become sufficiently established to support an inferred physical state.

​

​

From Prediction to Relation Intelligence

​

ORAYA can potentially help organisations answer questions such as:

• When has an early warning become trustworthy?

• Which prediction should be acted upon now?

• Which relation needs more evidence?

• Which model output should be withheld?

• Which previously reliable relation has become invalid under changed conditions?

• Which apparently strong relationship fails when tested against hidden contexts or decoys?

• What is the earliest stage at which a reliable decision can be made?

The opportunity is not necessarily to replace existing AI.

ORAYA may make the AI and sensor data an organisation already owns more valuable.

​

Potential ORAYA outputs include:

Admission maps

Abstention zones

First-admission surfaces

Never-admissible states

Context-sensitive trust boundaries

False-admission audits

Relation cards

Evidence-governance reports

​

Rocky Latchman

Founder, President and Inventor

​

Rocky Latchman is the founder and president of the PALA Institute for Spiral Cognition, the inventor and conceptual architect of ORAYA — Origin Relational Authentic Yield Arising, and the originator of the Latchman Admission Field.

His work develops system-level frameworks at the intersection of applied mathematics, physical sensing, artificial intelligence, cognition and engineering.

A central theme of Latchman's work is that relationships, conditions and boundaries can matter as much as the objects being measured.

ORAYA represents the most experimentally developed expression of this approach.

Latchman's central insight was that a predictive system should not automatically be entitled to assert every endpoint it can numerically estimate.

Instead, the relation itself should become an object of mathematical scrutiny:

When has the available evidence earned the right to support the claim?

​

From that principle developed ORAYA's architecture of relation admission, abstention, first trustworthy onset and context-sensitive relation intelligence.

ORAYA has now progressed through mathematical development, multiple public physical-data domains, blind testing, whole-context holdouts and external-reviewer attestation.

​

Patent Pending

Systems and Methods for Relation-Admission Analysis of Physical Sensor Data

​

ORAYA and its relation-admission architecture are the subject of a U.S. provisional patent application filed with the United States Patent and Trademark Office.

First Named Inventor: Rocky Latchman

U.S. Patent Pending

The patent filing forms part of PALA's programme to protect ORAYA's proprietary mathematical and technical architecture while developing scientific, industrial, licensing and commercial partnerships.

​

PALA Institute for Spiral Cognition

Research at the point where mathematics, intelligence and physical reality meet.

​

PALA is an independent Swiss research institute developing original frameworks across applied mathematics, physical sensing, artificial intelligence, cognition and engineering.

Our work begins from a simple principle:

A result should be understood through the conditions and relations that make it valid.

PALA develops ideas through a continuing research cycle:

Observation → Relation → Condition → Model → Experiment → Evaluation → Repair → New Observation

​

This is the spirit of spiral cognition: knowledge progresses not through one final assertion, but through increasingly disciplined cycles of formulation, testing, correction and extension.

ORAYA is currently PALA's principal applied relation-intelligence programme.

PALA's broader research architecture includes work in conditional systems, relation-centred mathematics, physics, cognition, abstraction and responsible technological design.

​

Why Partner with PALA?

​

ORAYA has moved beyond an initial theoretical proposal.

It now combines:

A named invention

A named mathematical object

Patent-pending status

An implemented architecture

Multiple public physical-data domains

Blind-testing infrastructure

Whole-context outcome-blind evaluation

External-reviewer attestation

A measurable advantage over its fixed-rule predecessor

The next major frontier is privately controlled industrial validation.

For organisations with difficult sensor, prediction or reliability problems, this creates an opportunity to test ORAYA in real operating environments.

Research Partnership

Work with PALA to investigate the mathematics, science and next generation of relation intelligence.

​

Industrial Validation

Apply ORAYA to privately controlled sensor data and a real operational problem.

Test whether relation admission, abstention and first-trustworthy onset produce measurable value in your domain.

​

Licensing and Co-Development

Explore integrating ORAYA's relation-intelligence architecture into existing products, platforms, monitoring systems or AI infrastructure.

​

Strategic Development and Investment

Support the transition of ORAYA from externally tested research architecture to scalable industrial technology.

Early strategic organisations may have the opportunity — subject to agreement and successful validation — to become associated with ORAYA's first industrial deployments and domain validation programmes.

Potential relationships may include:

Founding Industrial Validation Partner

Founding Research Partner

Strategic Development Partner

First Domain Deployment Collaborator

Investment or Research Sponsor

​

Be Part of the Next Stage of ORAYA

The next frontier in physical AI is not simply producing more predictions.

It is knowing:

when evidence becomes trustworthy,

when a relation should be admitted,

when a decision should wait,

and when the intelligent response is to abstain.

ORAYA is being built for that frontier.

​

PALA Institute for Spiral Cognition

Switzerland

Research · Mathematics · Physical AI · Cognition · Engineering

ORAYA — Origin Relational Authentic Yield Arising

U.S. Patent Pending

© PALA Institute for Spiral Cognition