model predictive control
PulseAugur coverage of model predictive control — every cluster mentioning model predictive control across labs, papers, and developer communities, ranked by signal.
12 day(s) with sentiment data
-
New DPC method offers deterministic safety guarantees for control systems
Researchers have developed a novel method for Differentiable Predictive Control (DPC) that provides deterministic feasibility guarantees, a critical aspect for safe control systems. This approach leverages topological a…
-
New CoCoNav framework enhances robot navigation safety in crowds
Researchers have developed CoCoNav, a new framework for robot navigation in crowded environments that enhances safety and efficiency. This system uses online conformal calibration to adapt trajectory-error bounds, allow…
-
AI framework tackles airport traffic congestion, cutting queues by up to 30%
Researchers have developed a computational framework inspired by QUBO to diagnose and optimize traffic flow in airport landside areas. This model, tested using data from Shanghai Pudong and Hangzhou Xiaoshan Internation…
-
PhyLatent improves JEPA world models by addressing physical state representation failures
Researchers have introduced PhyLatent, a novel training objective designed to enhance Joint Embedding Predictive Architecture (JEPA) world models. This new method aims to prevent specific failure modes in these models, …
-
New research proposes safety-gated LLM control for industrial systems
A new research paper introduces a safety-gated agentic supervisory control system designed to enhance the reliability of large language models (LLMs) in industrial control applications. The system incorporates a rule-ba…
-
New HOBA framework enhances online advertising bidding with hierarchical RL
Researchers have developed HOBA, a novel hierarchical reinforcement learning framework designed to improve online advertising bidding systems. This system decouples strategic reasoning, model selection, and bid executio…
-
Research paper details trade-offs in AI models for flow control
A new research paper explores the trade-offs between model compression and forecasting accuracy in data-driven reduced-order models for active flow control. The study compares Proper Orthogonal Decomposition (POD) with …
-
Multi-horizon consistency impacts video prediction geometry
Researchers have investigated the impact of multi-horizon latent consistency, a training parameter in video prediction and world models, on transition geometry. Their study, using Moving-MNIST as a primary test case, fo…
-
New RLxF framework challenges model uncertainty for AI safety
A new research paper proposes the RLxF (Reinforcement Learning from World Feedback) framework, challenging the reliance on internal model uncertainty as a risk signal in model-based reinforcement learning. The study dem…
-
RL vs. MPC for HVAC: Energy savings achieved, but RL faces comfort challenges
A new study published on arXiv compares the effectiveness of Reinforcement Learning (RL) and Model Predictive Control (MPC) for residential HVAC systems. Both methods demonstrated energy savings compared to traditional …
-
Pick-to-Learn methodology calibrates flight control policy using minimal scenarios
This paper introduces the Pick-to-Learn (P2L) methodology for calibrating Model Predictive Control (MPC) policies, demonstrated through an aircraft navigation problem. The P2L procedure identified two key wind scenarios…
-
OptCar adapts generalist models for high-speed off-road vehicle control
Researchers have developed OptCar, a method to adapt generalist vehicle models for high-speed, off-road autonomous control. OptCar uses a history-conditioned dynamics adaptation module and fine-tunes models with limited…
-
AI-powered digital twins enhance brain tumor prediction and treatment
Researchers have developed an AI-augmented adaptive digital twin framework to predict brain tumor evolution and optimize treatment schedules. This framework integrates a reaction-diffusion model with a 3D residual learn…
-
New hybrid framework enhances deformable object simulation
Researchers have developed Physics-Guided Residual Dynamics (PGRD), a novel simulation framework for deformable objects. This hybrid approach integrates a physics-based spring-mass simulator with a neural network that l…
-
New Physics-Guided Dynamics Model Enhances Deformable Object Simulation for Robotics
Researchers have developed Physics-Guided Residual Dynamics (PGRD), a novel simulation framework designed to improve the accuracy of predicting deformable object dynamics for robotics. PGRD integrates a physics-based sp…
-
New RLVR method fine-tunes reasoning models for energy storage control
Researchers have developed a novel method called Verifier-Based Reinforcement Fine-Tuning (RLVR) to adapt open-weight reasoning models for complex tasks like thermal energy storage control. This technique uses dynamic p…
-
New ADMM algorithm accelerates scenario-based model predictive control
Researchers have developed a novel learning-accelerated Alternating Direction Method of Multipliers (ADMM) algorithm to significantly speed up scenario-based model predictive control (SBMPC). This method reformulates SB…
-
New theory challenges prediction error as sole metric for latent world models
A new paper proposes a control theory framework for latent world models, challenging the assumption that minimizing prediction error directly leads to better control. The research argues that planners operate off the da…
-
MPC custody paper proposes post-quantum signature migration as key rotation
A new preprint on arXiv introduces a dual-gate MPC custody model that simplifies the transition to post-quantum signatures. This approach treats the switch to quantum-resistant signatures as a key rotation rather than a…
-
New TACTIC controller enhances whole-arm manipulation with tactile and vision data
Researchers have developed TACTIC, a new controller designed for whole-arm manipulation tasks that involve complex contact dynamics. This system integrates RGB-D vision, distributed tactile sensing, and a proximity repr…