Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HPRM: High-Performance Robotic Middleware for Intelligent Autonomous Systems

Jacky Kwok, Shulu Li, Marten Lohstroh, Edward A. Lee

TL;DR

HPRM introduces a deterministic, high-performance robotic middleware built on Lingua Franca to address nondeterminism and high IPC latency in large-scale sensor processing. It integrates an in-memory Plasma object store for zero-copy transfers, adaptive serialization to minimize data-copy overhead, and an eager protocol with real-time sockets to reduce handshake latency. The framework supports both centralized RTI-based coordination and decentralized, STP-offset-based coordination, enabling predictable timing or high availability as needed. Empirical results show up to 173x latency reduction over ROS2 for large-message broadcast and a 91.1% latency reduction in CARLA-based autonomous driving workloads, illustrating substantial practical impact for real-time autonomous systems.

Abstract

The rise of intelligent autonomous systems, especially in robotics and autonomous agents, has created a critical need for robust communication middleware that can ensure real-time processing of extensive sensor data. Current robotics middleware like Robot Operating System (ROS) 2 faces challenges with nondeterminism and high communication latency when dealing with large data across multiple subscribers on a multi-core compute platform. To address these issues, we present High-Performance Robotic Middleware (HPRM), built on top of the deterministic coordination language Lingua Franca (LF). HPRM employs optimizations including an in-memory object store for efficient zero-copy transfer of large payloads, adaptive serialization to minimize serialization overhead, and an eager protocol with real-time sockets to reduce handshake latency. Benchmarks show HPRM achieves up to 173x lower latency than ROS2 when broadcasting large messages to multiple nodes. We then demonstrate the benefits of HPRM by integrating it with the CARLA simulator and running reinforcement learning agents along with object detection workloads. In the CARLA autonomous driving application, HPRM attains 91.1% lower latency than ROS2. The deterministic coordination semantics of HPRM, combined with its optimized IPC mechanisms, enable efficient and predictable real-time communication for intelligent autonomous systems.

HPRM: High-Performance Robotic Middleware for Intelligent Autonomous Systems

TL;DR

HPRM introduces a deterministic, high-performance robotic middleware built on Lingua Franca to address nondeterminism and high IPC latency in large-scale sensor processing. It integrates an in-memory Plasma object store for zero-copy transfers, adaptive serialization to minimize data-copy overhead, and an eager protocol with real-time sockets to reduce handshake latency. The framework supports both centralized RTI-based coordination and decentralized, STP-offset-based coordination, enabling predictable timing or high availability as needed. Empirical results show up to 173x latency reduction over ROS2 for large-message broadcast and a 91.1% latency reduction in CARLA-based autonomous driving workloads, illustrating substantial practical impact for real-time autonomous systems.

Abstract

The rise of intelligent autonomous systems, especially in robotics and autonomous agents, has created a critical need for robust communication middleware that can ensure real-time processing of extensive sensor data. Current robotics middleware like Robot Operating System (ROS) 2 faces challenges with nondeterminism and high communication latency when dealing with large data across multiple subscribers on a multi-core compute platform. To address these issues, we present High-Performance Robotic Middleware (HPRM), built on top of the deterministic coordination language Lingua Franca (LF). HPRM employs optimizations including an in-memory object store for efficient zero-copy transfer of large payloads, adaptive serialization to minimize serialization overhead, and an eager protocol with real-time sockets to reduce handshake latency. Benchmarks show HPRM achieves up to 173x lower latency than ROS2 when broadcasting large messages to multiple nodes. We then demonstrate the benefits of HPRM by integrating it with the CARLA simulator and running reinforcement learning agents along with object detection workloads. In the CARLA autonomous driving application, HPRM attains 91.1% lower latency than ROS2. The deterministic coordination semantics of HPRM, combined with its optimized IPC mechanisms, enable efficient and predictable real-time communication for intelligent autonomous systems.

Paper Structure

This paper contains 18 sections, 1 equation, 6 figures.

Table of Contents

  1. Introduction
  2. Motivation and Requirements
  3. Motivation
  4. ROS2 vs. Lingua Franca
  5. ROS2
  6. Lingua Franca
  7. Coordination and Optimizations
  8. Centralized Coordination
  9. Decentralized Coordination
  10. Optimizations
  11. In-memory Object Store
  12. Adaptive Serialization
  13. Eager Protocol & Real-Time Sockets
  14. Evaluation
  15. Mean Latency
  16. ...and 3 more sections

Figures (6)

  • Figure 1: Architecture of in-memory object stores
  • Figure 2: Comparison of delay in writing NumPy arrays to shared memory
  • Figure 3: The serialization and deserialization throughput of ROS2 Numpy Package, in-band serialization, and out-of-band serialization
  • Figure 4: Mean latency of broadcast and gather on 4 nodes with different object sizes
  • Figure 5: Dataflow diagram of the autonomous driving application
  • ...and 1 more figures