model predictive control
PulseAugur coverage of model predictive control — every cluster mentioning model predictive control across labs, papers, and developer communities, ranked by signal.
9 day(s) with sentiment data
-
Akai, Moog, and Native Instruments parent company opens products to modders
InMusic, the parent company of music hardware brands like Akai, Moog, and Native Instruments, is opening up its products to the modding community. CEO Jack O'Donnell announced plans to provide a "toolbox" and inner work…
-
ZipDrift enhances short-horizon MPC with multimodal cost generation
Researchers have developed ZipDrift, a novel method for generating multimodal cost corrections in short-horizon model predictive control (MPC). This approach addresses the limitations of traditional MPC, which can be my…
-
Hybrid MPC and RL framework improves autonomous truck highway driving
Researchers have developed a hybrid model predictive control (MPC) and reinforcement learning (RL) framework for autonomous trucks on highways. This approach uses an RL model to estimate the terminal cost for a finite-h…
-
AI architecture SRLBC improves control for complex systems like nuclear reactors
Researchers have developed a novel architecture called Structured Representation Learning for Behavior Cloning (SRLBC) to improve the performance of AI controllers in complex dynamical systems. This method addresses cha…
-
Hybrid MPC and RL framework enhances autonomous truck highway driving
Researchers have developed a hybrid framework combining Model Predictive Control (MPC) and Reinforcement Learning (RL) to improve autonomous truck driving on highways. This approach uses an RL model to estimate the term…
-
Transformer Theories Tested: No Single Model Explains Capabilities
Researchers have evaluated seven different theories designed to explain the capabilities and limitations of transformer architectures. By translating these theories into a common matrix of predictions across various com…
-
Neural networks approximate robust model predictive control for reduced computation
Researchers have developed a new framework that uses neural networks to approximate model predictive control (MPC), a high-performance control method. This approach aims to reduce the significant online computational co…
-
Neural Networks Provably Inherit Safety Guarantees from Model Predictive Control
Researchers have developed a method to create neural network controllers that provably maintain the safety and stability guarantees of robust model predictive control (MPC). This approach treats the neural network's app…
-
New algorithm bridges POMDP theory and deep RL practice
Researchers have developed a novel end-to-end algorithm for model-based reinforcement learning in partially observable environments. This algorithm addresses the challenge of inferring hidden states, learning latent dyn…
-
Multi-agent RL method enhances UAV pathfinding in stochastic environments
Researchers have developed a novel multi-agent reinforcement learning approach for UAV pathfinding in stochastic environments. This method utilizes a centralized training with decentralized execution strategy, enhanced …
-
Musicians embrace custom firmware for synths and MPCs, favoring open-source control
Musicians are embracing custom firmware for their synthesizers and MPCs, a trend that highlights the benefits of open-source hardware. This movement allows users to modify, customize, and extend the lifespan of their mu…
-
New D2OC method enables efficient nonlinear multi-agent spatial coverage
Researchers have developed a nonlinear extension of Density-Driven Optimal Control (D2OC) for multi-agent spatial coverage. This new method, D2OC, drives agent distributions toward a desired density using a Wasserstein-…
-
New research advances robot world models for efficient planning and control · 10 sources tracked
Recent research explores advancements in world action models (WAMs) for robotics and agent planning. WAM-Cache and WAMJET focus on accelerating inference by optimizing key-value reuse and employing agentic harnesses for…
-
Neural network approach estimates RC thermal building model parameters for MPC
A new research paper introduces the "Estimator from Scratch" approach for estimating parameters in RC thermal building models, which are crucial for energy-efficient model predictive control (MPC). This method embeds ph…
-
Robotic peg insertion enhanced by learned contact dynamics model
Researchers have developed a novel graph neural network model capable of learning contact dynamics for robotic manipulation tasks like peg insertion. This model, trained through self-supervision using only touch and for…
-
New backdoor attack targets AI world models for control systems
Researchers have identified a new supply-chain vulnerability in pretrained world models, which are used as simulators for control tasks. An adversary can embed a backdoor in a model checkpoint, allowing them to hijack d…
-
New variable horizon method enhances multi-drone collision avoidance
Researchers have developed a novel conflict-predictive variable horizon approach for multi-drone distributed model predictive control. This method allows each drone to dynamically adjust its prediction horizon based on …
-
New transfer learning model enhances building control performance
Researchers have developed a novel zero-shot transfer learning approach for model predictive control (MPC) in buildings. This method utilizes generalized models pretrained on excitation-based operational data, which pur…
-
New research compares control strategies for sustainable hydrogen supply chains
Researchers have developed and compared four control approaches for optimizing renewable-powered hydrogen supply chains. Model Predictive Control (MPC) demonstrated the highest economic performance by utilizing short-te…
-
New MPC framework uses Gaussian Processes for robust control
Researchers have developed a new model predictive control (MPC) framework designed for uncertain nonlinear systems. This framework utilizes Gaussian Processes (GPs) to learn system dynamics from noisy measurements, inco…