ACC 2021
153 papers
The 2021 American Control Conference (ACC) featured 146 papers with a strong emphasis on decentralized optimization, safety-critical control, and distributed systems. Key research themes included privacy-preserving distributed optimization, Byzantine-resilient stochastic gradient descent, and decentralized stability certificates for inverter-based power grids. Several studies addressed the reduction of conservatism in safety-critical control and the development of robust optimization methods for dynamic network topologies.
The year’s contributions also spanned diverse applications, ranging from identifying cancer cell population dynamics for drug scheduling to data-driven anomaly detection in electricity markets. Notable work explored hierarchical constraint satisfaction for autonomous systems and contact pose identification in robotic assembly. Prolific authors contributing to these advancements included Amrit Singh Bedi, Bike Zhang, Koushil Sreenath, and Claire J. Tomlin. Collectively, the research highlighted a trend toward integrating learning-based methods with classical control, focusing on non-asymptotic convergence analysis and automata-based specifications for non-Markovian rewards.
Top Research Directions
Research directions most represented by ACC 2021 papers.
Top Authors
Impactful Papers
Papers from ACC 2021 that shaped research clusters.
Foundational
Joint state and dynamics estimation with high-gain observers and Gaussian process models
foundational
Learning Discrete Time Flow Maps
Influential
Frontiers in Scalable Distributed Control: SLS, MPC, and Beyond
influential
Convex Control Synthesis via Closed Loop
Approximate Stochastic Reachability for High Dimensional Systems
influential
Scaling Kernel Methods with Random Fouri
Approximate Stochastic Reachability for High Dimensional Systems
influential
The Dimensionality Trade-off in Random F
On the Regret Analysis of Online LQR Control with Predictions
influential
Dynamic Regret Minimization for Online C
On Occupation Kernels, Liouville Operators, and Dynamic Mode Decomposition
influential
Regularizing Unbounded Operators for Sta
Poisoning Attacks against Data-Driven Control Methods
influential
Stealthy False Data Injection in Cyber-P
Robust Asynchronous and Network-Independent Cooperative Learning
influential
Geometric Belief Aggregation in Decentra
Adaptive Scheduling for Machine Learning Tasks over Networks
influential
Tunable Communication Thresholds for Dis
Advancing Trajectory Optimization with Approximate Inference: Exploration, Covariance Control and Ad
influential
Sequential Decision Making as Probabilis
Papers
« prev Page 4/4