Author name cluster

Edward A. Lee

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

ICRA Conference 2025 Conference Paper

HPRM: High-Performance Robotic Middleware for Intelligent Autonomous Systems

Jacky Kwok
Shulu Li
Marten Lohstroh
Edward A. Lee

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 114x 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. Code and videos can be found on our project page: https://hprm-robotics.github.io/HPRM

Details

ICLR Conference 2020 Conference Paper

Learning Heuristics for Quantified Boolean Formulas through Reinforcement Learning

Gil Lederman
Markus N. Rabe
Sanjit A. Seshia
Edward A. Lee

We demonstrate how to learn efficient heuristics for automated reasoning algorithms for quantified Boolean formulas through deep reinforcement learning. We focus on a backtracking search algorithm, which can already solve formulas of impressive size - up to hundreds of thousands of variables. The main challenge is to find a representation of these formulas that lends itself to making predictions in a scalable way. For a family of challenging problems, we learned a heuristic that solves significantly more formulas compared to the existing handwritten heuristics.

Details

TCS Journal 2015 Journal Article

The fixed-point theory of strictly causal functions

Eleftherios Matsikoudis
Edward A. Lee

We ask whether strictly causal components form well defined systems when arranged in feedback configurations. The standard interpretation for such configurations induces a fixed-point constraint on the function modeling the component involved. We define strictly causal functions formally, and show that the corresponding fixed-point problem does not always have a well defined solution. We examine the relationship between these functions and the functions that are strictly contracting with respect to a generalized distance function on signals, and argue that these strictly contracting functions are actually the functions that one ought to be interested in. We prove a constructive fixed-point theorem for these functions, introduce a corresponding induction principle, and study the related convergence process.

Details DOI

TCS Journal 2008 Journal Article

CPO semantics of timed interactive actor networks

Xiaojun Liu
Edward A. Lee

We give a denotational framework for composing interactive components into closed or open systems and show how to adapt classical domain-theoretic approaches to open systems and to timed systems. For timed systems, prior approaches are based on temporal logics, automata theory, or metric spaces. In this paper, we base the semantics on a CPO with a prefix order, as has been done previously for untimed systems. We show that existence and uniqueness of behaviors are ensured by continuity with respect to this prefix order. Existence and uniqueness of behaviors, however, do not imply that a composition of components yields a useful behavior. The unique behavior could be empty or smaller than expected. We define liveness and show that appropriately defined causality conditions ensure liveness and freedom from Zeno conditions. In our formulation, causality does not require a metric and can embrace a wide variety of models of time.

Details DOI