ACC 2025

116 papers
The 2025 American Control Conference (ACC) featured 76 papers focusing on diverse areas of control theory and its applications. Prominent research directions included navigating the Pareto frontier of algorithmic fairness and developing preference-conditioned policies for multi-objective reinforcement learning. Technical contributions spanned gradient flow dynamics for low-rank matrix factorization, H-infinity robust dynamic decoupling for precision motion systems, and the use of diffusion models for the manifold-guided stabilization of nonlinear dynamical systems. Additional research explored multi-agent causal dynamics for temporally extended tasks and strategies for achieving cooperation compliance among human drivers in mixed traffic environments. Notable contributors to this year's proceedings included Sanjay Lall, Zhe Xu, Jinming Xu, Peter Seiler, Ufuk Topcu, Marco Pavone, and Simone D'Amico. Other specific studies addressed buffer occupancy in bittide systems and feasibility-oriented algorithms for radial reconfiguration in distribution networks. These works reflect an emphasis on integrating machine learning techniques, such as reward machine inference and diffusion models, with classical control and optimization frameworks like bilevel optimization and gradient flow.

Top Research Directions

Research directions most represented by ACC 2025 papers.

Generalizing Provable Control to Nonlinear Systems 8
Accelerating Trajectory Optimization for Real-Time Robotic Control 3
Bridging the Simulation to Reality Transfer Gap 3
Control Barrier Functions for Safe Autonomous Systems 2
Optimization Based Safety Filtering via Control Barrier Functions 2
Formal Stability Guarantees for Neural Control Systems 2
Eco-Routing for Emission and Exposure Reduction 1
Evaluating Automated Blindspot Discovery for Concurrent Failures 1
Transferring Safety Guarantees from Abstract Models to Concrete Systems 1
Non-Convex and Disjoint Prediction Regions 1
Look-Ahead Mechanisms for Predictive Path Tracking 1
Equitable Alternatives to Regressive Congestion Pricing 1
Overcoming Noise Sensitivity in Symbolic Discovery 1
Reduced Order Models for Real Time Locomotion Planning 1
Noise Amplification in Numerical Differentiation 1

Top Authors

AuthorPapers
John Subosits5
Marco Pavone3
Peter Seiler3
Sanjay Lall3
ufuk topcu3
Zhe Xu3
Ahmad Tawaha2
Calin Belta2
Colin Jones2
David Fridovich-Keil2
Guannan Qu2
Guojian Zhan2
Kaidi Yang2
Mohamed El Mistiri2
Shahriar Talebi2
Shengbo Eben Li2
Tamer Basar2
Thomas Lew2
Yujie Tang2
Aaron Ames1

Impactful Papers

Papers from ACC 2025 that shaped research clusters.

Influential

Papers

Title
Stability Margins of Neural Network Controllers
State Space Models as Foundation Models: A Control Theoretic Overview
Stochastic MPC with focus on probabilistic constraints with application to wind turbine control
Structure-Exploiting Reinforcement Learning for Networked Systems
SwarmCVT: Centroidal Voronoi Tessellation-Based Path Planning for Very-Large-Scale Robotics
Symbolic Regression on Sparse and Noisy Data with Gaussian Processes
Task Decomposition for Learning Advanced Driving Skills
Time-reversal solution of BSDEs in stochastic optimal control: a linear quadratic study
Towards Achieving Cooperation Compliance of Human Drivers in Mixed Traffic
Trajectory-Based Automata Learning for Offline Reinforcement Learning
Transient Control of Linear Multi-Agent Systems with Leader-Follower Configuration
Using Large Language Models to Automate and Expedite Reinforcement Learning with Reward Machine
Variance-Reduced Gradient Estimator for Nonconvex Zeroth-Order Distributed Optimization
YourMove: A System Identification and Hybrid Model Predictive Control Personalized mHealth Intervention for Physical Activity
Zero-Order Control Barrier Functions for Sampled-Data Systems with State and Input Dependent Safety Constraints
Zeroth-Order Feedback Optimization in Multi-Agent Systems: Tackling Coupled Constraints

« prev Page 3/3