Infrastructure
In today’s interconnected world of cloud, servers, and devices, Uli SDK organizes computing resources into logical subsystems and nodes. As depicted in the diagram below, a subsystem comprises one or more nodes, and each node hosts one or more computational applications (comp apps) that implement related services. These services communicate with one another using message-based interactions.
Subsystems, nodes, and comp apps are each assigned unique identifiers within their respective contexts. Messages exchanged between servers and clients are addressed using tuples in the format (Subsystem ID, Node ID, Comp ID), which specify both the source and destination of the communication.
The Subsystem ID and Node ID can be dynamically assigned. Similar to how a DHCP server assigns IP addresses, the Uli SDK’s Id Allocator service assigns unique IDs to requesting subsystems. Once a Subsystem ID is assigned, the Subsystem Manager service within that subsystem allocates IDs to the requesting nodes.>
Messages are serialized for efficiency using Cap’n Proto (see capnproto.org). The serialized messages communicate between servers and clients via local socket and UDP unicast and multicast packets. Messages can also be assigned to specific partitions so that only the servers and clients of the same partition can communicate.
Each node implements a Node Manager service that handles message routing. It dynamically learns routing information from the messages it processes, enabling it to forward messages to their destinations without requiring IP addresses for communication.
Services
For each service, Uli SDK defines and implements input/output messages, internal events (events within the comp app), a state machine, and configurable parameters. Related services are grouped together into computational applications (Comp Apps).
Each Comp App must include the Service Register and Status Reporter services. The Service Register enables the discovery of services within the Comp App, while the Status Reporter provides detailed information on the status of those services.
Resources
Resources are digital entities within the subsystem, encompassing video sources, web sockets, drivers, manipulators, machine learning (ML) models, and meta characters. Clients can access these resources via the Resource Discovery service.
Agents
Agents are computing modules that perform mission-critical tasks. They are discoverable via the subsystem's Agent Discovery service, allowing clients to query each agent's capabilities, configurations, and status, and to request exclusive control over them.
Agents serve as essential resources for AI Agents within LLM frameworks. They provide the tools necessary for task execution, deliver capability and mission-readiness data that support AI-driven workflow reasoning, and supply control parameters and feedback to help AI Agents learn and optimize processes.
Data Topics
Data Topics refer to the categories of data published by the services. These can be discovered through the Data Topic Discovery service of the subsystem. Clients can subscribe to the Data Topics of interest.
Transforms
Transforms represent the quaternion conversion between two coordinate frames. They are provided by the Transform Reporter services, which can be located via the Transform Discovery service of the subsystem. Clients can query the Transform Reporter services to obtain the transform between specific coordinate frames.
Mission Execution
The mission is carried out through agents that perform essential tasks. The Uli SDK offers a unified interface for discovering, configuring, executing, and monitoring the status of these agents.
Data Access
The Uli SDK supports Data Topic Discovery, allowing applications within the subsystem to register their published data topics. Through a unified interface, clients can discover and subscribe to these data topics.
Data topics can be visualized using integrated Google MESOP web apps or can be input into Data Ontology Tools to generate additional knowledge for Large Language Models.
Safety
The Uli SDK includes a Safety Service that monitors user-initiated emergency stop (e-stop) events and the health status of services within applications of the subsystem. When an emergency condition is detected, it is escalated and propagated to safety-critical functions.
Additionally, application message transports can be partitioned to separate communications between safety-critical and general-purpose functions, enhancing the overall safety of the system.
Security
To safeguard against cyberattacks, the Uli SDK employs secure messaging and authentication between services and clients. For inter-node communications, it uses Datagram Transport Layer Security (DTLS) to ensure secure data transmission. For mission control and data access, certificate-based authentication is used to verify clients and manage their access permissions.
Integration
The Uli SDK is responsible for the orange-outlined blocks in the diagram below. These top-level blocks—Operator Controllers, Subsystem Managers, Node Managers, Comp Apps, and Proxies—are composed of services that implement standard interfaces, to the blue-outlined blocks:
Tools
The Uli SDK offers a comprehensive suite of tools for code generation, building, and application deployment.
The SDK's code generation tools produce concise C++ code for record structures, messages, services, and Comp Apps. For example, the Uli-generated service code typically contains around 400 source lines of code (SLOC).
Uli SDK utilizes Google's highly efficient build tool, Bazel, which supports cross-building for both x86 and ARM 64-bit architectures, with or without CUDA support. It has been tested on Ubuntu 18.04, 20.04, 22.04, and 24.04, and supports x86_64, NVIDIA Jetson, Xavier, and AGX boards. Here are the Uli SDK supported OS and devices:
In addition, the SDK includes shell scripts to streamline the staging and deployment of applications across a network of devices
| OS | Device | Development Host (x86_64) |
| Jetpack 46 | Jetson Nano, Jetson Xavier NX | Ubuntu 18.04 |
| Jetpack 5 | Jetson Xavier NX, Jetson Orin | Ubuntu 20.04 |
| Jetpack 6 | Jetson Orin Nano, AGX Orin | Ubuntu 22.04 |
| Ubuntu 18.04 | x86_64 | Ubuntu 18.04 |
| Ubuntu 20.04 | x86_64 | Ubuntu 20.04 |
| Ubuntu 22.04 | x86_64 | Ubuntu 22.04 |
| Ubuntu 24.04 | x86_64 | Ubuntu 24.04 |
Highlights
1. Modular Open System Approach (MOSA) provides interoperability and flexibility of the system architecture.
2. Communication path partition that only servers and clients of the same partition can communicate.
3. Secure messaging and mutual authentication between services and clients.
4. Dynamic ID assignment for assets, computing modules, and applications.
5. Discovery of services, resources, data topics, agents, and transform reporters.
6. Data access authorization categorized into classified, controlled, and unclassified tiers.
7. Exclusive service control, ensuring controlled access to services.
8. Lifecycle management of services, including startup, runtime, and shutdown phases.
9. Service health reporting and monitoring to track Comp App status.
10. Emergency stop (e-stop) propagation to allow services to manage critical situations.
11. Support for various operational modes, including Standard, Training, Maintenance, and user-defined modes.
12. Integration with Google MESOP, ROS1, ROS2, Foxglove Studio, AI Agents, NVIDIA ISAAC Sim, and Omniverse.
13. Code generation tools that produce C++ code for record structures, messages, services, Comp Apps, and Python bindings.
14. Cross-build support for x86_64, NVIDIA Jetson Nano, Xavier, and AGX, as well as multiple Ubuntu versions (18.04, 20.04, 22.04, 24.04).
15. Comprehensive examples of Comp Apps and QT PySide6-based UI.
Advantages
The Uli SDK stands out for its secure and dynamic features, providing key advantages:
Safety: Monitoring the user e-stop event and the health conditions of the services. Partition message transports between safety-critical and general-purpose functions.
Security: Secure messaging and mutual authentication between services and clients.
Unified agent and data access interfaces: enable the discovery of agents and data topics, and support agent execution and data topic subscriptions. Data access is categorized into classified, controlled, and unclassified levels, with separate storage for each.
Quality: Quality assurance is embedded throughout the development lifecycle, with continuous review and validation to ensure objectives are met.
Cost-effectiveness: The Uli SDK is open-source for licensed customers, offering low overall costs.