IEEE P2874 Working Group

Spatial Web Protocol, Architecture and Governance

​

“Public imperative”

​

https://sagroups.ieee.org/2874/

2024-06-06 - D3.1.1

IEEE P2874 Scope

This specification defines

• Scope and key concepts of a Spatial Web framework,
• Relevant terms and definitions
• Generalized system protocol, architecture, and governance
• that together serve as a reference model for Spatial Web components, systems, services and specifications.

This reference model establishes

• the set of required modules and sub-modules and their minimum functions,
• the associated information content and the information models
• to be provided by a Standards-based compliant Spatial Web implementations

Spatial Web: agents in a cyber-physical ecosystem

Intelligent AGENTS with CREDENTIALS

performing ACTIVITIES discussed in CHANNELS,

on and in DOMAINS represented in HYPERSPACES,

fulfilling CONTRACTS with other AGENTS.

IEEE P2874 Spatial Web

1. Overview

2. Normative references

3. Definitions, acronyms

4. Spatial Web standards

5. Value for stakeholders

6. Conceptual model

7. Distributed computing

A. Compliance Components

B. Bibliography

5. Value for stakeholders

5.1. Guiding principles

5.3. Stakeholder perspectives

5.4. Application scenarios

6. Conceptual model

6.1. Overview

6.2. Space, time, hyperspace

6.3. Domain modeling and identity

6.4. Agents and activities

6.5 Credentials, norms, contracts

6.6. Spatial Web ontology

6.7. Query requirements

7. Distributed computing

7.1. Computing architectures

7.2. Spatial web nodes

7.3. HSTP conceptual model

7.4. Distributed computing use cases

4. Spatial Web standards

4.1. SW set of standards

• This document specifies requirements for a set of implementation standards
• For interoperability and governance of cyber-physical systems
• Including autonomous devices, applications, spatial content and operations.

4.2. Socio-technical standards

• This is a Socio-technical standard
• Includes stakeholder perspectives linked to technical requirements
• Basis for AI governance

4.3. System design

• This standard provides an architectural design of the Spatial Web
• Artifact from an iterative system design process

4.3. System architecting and implementation

5. Value for stakeholders

5.1. Guiding principles

5.2. Stakeholder perspectives

5.3. Application scenarios

​

​

5.1 Guiding principles

• Spatiality
• Best organizing principle for most information is as a set of locations in hyperspace.
• Interoperability
• Searchability, viewability, interaction, transaction, and transportation of any asset or user within or across any spaces.
• Governance / Trust
• Real-time, permission-driven validation of all users, assets, and spaces and their interactions
• Privacy
• Individual control, trust, and security with cryptographically - secured and decentrally-stored digital identity enables “trustless” transactions with anonymity and auditability.
• Ownership
• Users can own their data and digital property and choose with whom they share this data.
• Security
• Secure data collection, transmission, and storage enables interactions and transactions with virtual and physical assets
• Responsibility
• Creating technology guided by upholding ethical principles of inclusivity, transparency and cooperation with the goal of creating a better world for all humanity.

5.2 Stakeholder perspectives

​

​

5.3 Application Scenarios

​

10

Application scenarios range over geographic and societal scales

6. Conceptual model

6.1. Overview

6.2. Space, time, hyperspace

6.3. Domains and identity

6.4. Agents and activities

6.5 Credentials, norms, contracts

6.6. Spatial Web ontology

6.7. Query requirements

6.2. Space, time, hyperspace

​

(Mathematical foundations)

hyperspace - generalized concept of space

A hyperspace is “a set of elements with paths between them”�(equiv. to a category!)

Hyperspace includes �geospace, abstract spaces, high-dim. vector spaces,�cellular spaces,�hypergraphs,�other higher-order graphs, …

“Paths” can be as general as processes / programs !

​

Embedding geographic space in hyperspace

6.3 Domain modeling and identity.

Domain: Entities with identity through time endowed with rights and credentials. DOMAIN is an ENTITY in the Spatial Web ontology

Domain Authority: An entity that is credentialed to have the ability to define within a Domain the norms and terms under which contracts are created governing: agents, actions and credentials within that Domain

Universal Domain Graph (UDG): Distributed metagraph containing all relationship information between all known Domains in the Spatial Web.

Identifier: A unique attribute or a combination of attributes that distinctly characterizes and distinguishes each {change Domain to Entity} in the Spatial Web network. This identifier is essential for recognizing, tracking, and interacting with entities consistently and unambiguously. Each entity in the Spatial Web has a unique Spatial Web Identifier (SWID).

Identity: A respect in which things are equivalent, or a relation of equivalence. Identity may be relative to a set of shared properties (qualitative), or absolute (numerical identity). In qualitative terms, a Domain may have more than one identity (belong to more than one equivalence class), and several Domains can have the same identity.

14

6.3 Domain types

15

6.3 Universal Domain Graph

The UDG is a Global Commons and a public utility that is

accessible and usable by all Spatial Web participants enabling:

​

Interoperability: By recognizing how Domains interact, the UDG

may facilitate the seamless exchange of information between different

Domains.

​

Complex Searches: With nested, hierarchical, and heterarchical relationships defined, searches may be more sophisticated, allowing users to find information that is contextually relevant across multiple dimensions and scales.

​

Modeling and Analysis: Understanding these relationships is essential for modeling the real world in a way that reflects its complexity, enabling better analysis, forecasting, and decision-making.

​

Adaptability and Scalability: As entities move and relationships change, the UDG’s structure must adapt. Recognizing these dynamic relationships ensures that the UDG remains relevant and accurate.

​

Governance: Understanding the intricate relationships between Domains enhances governance by promoting a more comprehensive, coherent, and effective approach to policy-making, planning, and decision-making. It enables governments to navigate the complex challenges of our interconnected world more successfully and promote the well-being of their constituents and the broader society.

​

6.4 Agents and Activities.

​

Agent

• Domain that senses and responds to its environment
• maintains a model of its environment
• takes actions to achieve its goals.

​

Activity

• An action or set of actions performed by an Agent.
• Performed on, in, with, or to a Domain

Relationship:

Agents interact with and influence their environment via sensing and actuation activities.

HSML Activities supports the encoding, interpretation, execution, & governance of those activities.

6.4 Agents

​

Agent Capabilities

• Can sense the environment
• Has a model of the environment
• Can act through actuation
• Can take action to achieve goals

​

Functions & Capabilities within Spatial Web

• Intelligent, proactive, adaptive, collaborative
• Information gathering and analysis
• Autonomous decisions-making and goal-setting
• Modify and influence entities in its environment

​

​

​

​

​

​

6.4 Agent Interactivity & Interoperability

​

Interaction Heterogeneity

​

Agent-Environment Interactions:

• Physical, virtual, digital, hybrid domains.
• Hybrid: digital twins, augmented or mixed realities, smart cities.

​

Agent-Domains Interactions:

• IoT devices, virtual or physical assets, databases, services.

​

Agent-Agent Interactions:

• Human-AI Interactions & Human-AIS Teaming
• Peer-2-Peer Networks
• Multi-agent Systems

​

Communication Complexity

• Low: simple queries or digital commands, fast time-scales, low-latencies.
• High: syntactically & semantically rich representations, multiplexed time-scales.

​

​

​

​

6.4 Agent Governance

AIRS framework is a standardized rating system for identifying and characterizing the various capabilities of autonomous intelligent agents.

​

It will serve as a foundation for the governance of autonomous intelligent agents.

• “Levels of Intelligence” will serve as a key criterion for agent agent classification
• Encompassess aspect of a ‘function-based’, ‘autonomy-based’, ‘performance-based’ classification system.
• Complements

​

​

​

6.4 Agents & Activities

​

Activities as the locus of action representation…

​

agent capabilities: with defined Activities, we can depict the range of capabilities or affordances of an agent.

​

agent planning & intentions: Activities represent an agent’s plans, goals, and intended actions.

​

agent collaboration & conflict avoidance: provide a structured way to represent and coordinate interactions or commitments among multiple agents.

​

Activities for governance:

​

Serves as a framework for encoding ethical guidelines and regulatory requirements into agent behavior.

Activity registry supports the discovery and understanding of agent capabilities.

​

Record of chain-of-responsibility, chain-of-failure.

​

​

​

​

​

​

​

​

​

​

​

Ag

6.5 Credentials, Norms, Contracts

• Credential

Information that an entity possesses that allows it to make verifiable claims

• Contract

A binding agreement between two parties, especially enforceable by law, or a similar internal agreement wholly within an organization

• Norm

A standard, or principle of right action, binding upon the members of a group and serving to guide, control, or regulate behavior.

22

Discussion of how normative concepts are represented in HSML

Safety and trustworthiness of SW agents via governance

As discussed previously, safety/trust is enabled by governance in two senses:

• self-regulation of autonomous agents
• regulation of agent Activities via norms, contracts, and credentials, and the ability of agents to reason about them

​

Governance via Contracts, Activities and Credentials:

• [Activity] represents a possible action in terms of its preconditions and consequences
• Those conditions may include possession of a [Credential] by relevant actors
• [Contract] instantiates an [Activity] with free variables bound to particular agents, Domains, spaces, times, etc.

​

​

​

​

​

Credential may hold several types of claims

24

Credentials and claims: decentralized registry management

25

Norms and normativity

Norms is a broad category that subsumes rules, laws, and other practices or standards binding on specific groups or Domains.

• Frameworks exist for formally describing and computing normative statuses i.e. automated governance for autonomous agents:
• HSML Activities, Contracts and Credentials play a similar role, and could be implemented via these systems
• Ethical & legal codes: describe what ought to or must occur
• Activities: describe what must occur in order to fulfill a goal/intention

​

​

​

Norms organized by Domains

A Domain Authority may define rules that apply to Activities occurring within its governed Domain(s). These may represent laws endorsed by an organization (for state actors) as well as ethical principles (cf. IEEE P7007), customs, etc.

​

Complete Activity validation will often depend simultaneously on norms from multiple hierarchical levels (e.g. Business < State < National)

• Rules themselves, or their enforcement, may come into conflict with rules imposed by a superordinate Domain.
• The SW architecture must be able to represent such conflicts, and Norms should be subject to the same kind of validation as Activities.

​

​

​

​

​

​

Norms as constraints on Activities

Given Activity conditions, there are at least two ways to provide governance on which sorts of Activities can be executed in a Domain:

​

• Governance via Activity filtering:
• Greenlight, prohibit, or request changes to Activities based on Conditions
• This can be applied at the Schema or instance level
• Governance via Activity modulation:
• Create a new fine-grained Activity that is approved by the Domain Authority
• This can be combined with the above strategy
• Domain Authorities as well as executing agents can pre-specify which changes to Activities they will require and allow, respectively (negotiation)

​

​

​

​

​

Agent-Contract-Activity relationship diagram

6.6. Spatial Web ontology

6.7. Queries

• Bootstrapping/Context queries
• Activity queries
• Hyperspace range queries
• Abstract data type query
• Graph queries
• Semantic queries
• Vector queries
• Queries on HSML Entitites

​

31

6.7. Query examples

32

Hexagon cell index query

Multi-Coordinate Query

7. Distributed computing

7.1. Computing architectures

7.2. Spatial web nodes

7.3. HSTP conceptual model

7.4. Distributed computing use cases

7.1.1 Distributed Computing Continuum

34

7.2. Spatial Web Nodes

Computing machines connected to the internet, capable of exchanging HSTP messages.

35

• Client nodes
• Spatial Web browser
• Spatial Web application client
• Immersive experience
• Cloud and local server nodes.
• UDG node
• HSML content node
• Processing nodes
• Data warehouse
• AI Agent

• Edge nodes
• IoT gateway
• HSML content node
• AI agent.
• Physical devices
• Embedded node
• Moving platform with sensors and actuator; robots
• Stationary platform with sensors and actuators;

7.3. HSTP conceptual model

• HSTP: Hyper Space Transaction Protocol
• Coherent, decentralized, secure, and privacy-respecting communications
• Transaction focused; Transport agnostic

​

• HSTP Objects
• instances of a HSML Activity (Event)
• denotes desired HSML state change

​

• Distributed computing use cases
• HSTP messages between Spatial Web nodes

36

7.3.3.1 HSTP OPERATIONs

7.4. Distributed computing use cases with HSTP

• Exchange of HSTP messages between Spatial Web nodes

​

• Messages contain HSML

​

• Described using sequence diagrams

​

• Coherence of architecture demonstrated by use cases satisfying application scenarios

​

38

7.4.3. Use Case: Create and register a new DOMAIN

7.4.11. Use Case: Discover DOMAINS using SPACEs

7.4.14. Use Case: Transfer DOMAIN between DOMAINs

IEEE P2874 Spatial Web

1. Overview

2. Normative references

3. Definitions, acronyms

4. Spatial Web standards

5. Value for stakeholders

6. Conceptual model

7. Distributed computing

A. Compliance Components

B. Bibliography

5. Value for stakeholders

5.1. Guiding principles

5.3. Stakeholder perspectives

5.4. Application scenarios

6. Conceptual model

6.1. Overview

6.2. Space, time, hyperspace

6.3. Domain modeling and identity

6.4. Agents and activities

6.5 Credentials, norms, contracts

6.6. Spatial Web ontology

6.7. Query requirements

7. Distributed computing

7.1. Computing architectures

7.2. Spatial web nodes

7.3. HSTP conceptual model

7.4. Distributed computing use cases

Authors, Editors, Contributors

WG D2 comment contributors

• Geert Devos
• Flores Bakker
• Karl Friston
• Tom De Block
• Mikel Salazar
• Martin Laskowski

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

SWF Writing Team

• Gabriel Rene
• Dan Mapes
• Christine Perey
• Prasaanth Sridharan
• Capm Petersen
• Alec Tschantz
• Toby St Clere Smithe
• Mahault Albarracin
• Jacqueline Hynes

​

​

• Alex Kiefer
• Jason Fox
• Maxwell Ramstead
• Reese Plews
• Philippe Sayegh
• Sarah Manski
• Scott Carroll
• George Percivall