Learning-Based Multi-UAV Flocking Control With Limited Visual Field and Instinctive Repulsion

This article explores deep reinforcement learning (DRL) for the flocking control of unmanned aerial vehicle (UAV) swarms. The flocking control policy is trained using a centralized-learning-decentralized-execution (CTDE) paradigm, where a centralized critic network augmented with additional information about the entire UAV swarm is utilized...

Automatic enhancement of vascular configuration for self-healing concrete through reinforcement learning approach

Vascular self-healing concrete (SHC) has great potential to mitigate the environmental impact of the construction industry by increasing the durability of structures. Designing concrete with high initial mechanical properties by searching a specific arrangement of vascular structure is of great importance. Herein, an automatic optimization...

Generalized Model and Deep Reinforcement Learning-Based Evolutionary Method for Multitype Satellite Observation Scheduling

Multitype satellite observation, including optical observation satellites, synthetic aperture radar (SAR) satellites, and electromagnetic satellites, has become an important direction in integrated satellite applications due to its ability to cope with various complex situations. In the multitype satellite observation scheduling problem ...

Optimization Strategies for System Architecting Problems

System architecting is one of the first stages of the engineering problem-solving process. Pivotal decisions regarding the system's overall configuration are taken in this phase. Consequently, decision support tools like system architecture optimization are needed to effectively assess the architectural design space. However, system architecture...

Online Optimization of Gear Shift and Velocity for Eco-Driving using Adaptive Dynamic Programming

In this paper a learning-based optimization method for online gear shift and velocity control is presented to reduce the fuel consumption and improve the driving comfort in a car-following process. The continuous traction force and the discrete gear shift are optimized jointly to improve both the powertrain operation and the longitudinal...

Timeslot allocation for waiting list control: Tactical planning of orthopaedic surgeons at the Sint Maartenskliniek

Patients visiting a hospital for elective surgery often have multiple consultations with a surgeon before undergoing surgery. Hospitals discern between different types of consultations, and make a schedule allocating timeslots of outpatient department sessions to these different consultation types several weeks in advance. Changing the...

Machine Learning in Chemical Engineering: A Perspective

The transformation of the chemical industry to renewable energy and feedstock supply requires new paradigms for the design of flexible plants, (bio-)catalysts, and functional materials. Recent breakthroughs in machine learning (ML) provide unique opportunities, but only joint interdisciplinary research between the ML and chemical engineering ...

Fine-tuning deep RL with gradient-free optimization

Deep reinforcement learning makes it possible to train control policies that map high-dimensional observations to actions. These methods typically use gradient-based optimization techniques to enable relatively efficient learning, but are notoriously sensitive to hyperparameter choices and do not have good convergence properties. Gradient...

Resource-constrained Multi-agent Markov Decision Processes

Intelligent autonomous agents, designed to automate and simplify many aspects of our society, will increasingly be required to also interact with other agents autonomously. Where agents interact, they are likely to encounter resource constraints. For example, agents managing household appliances to optimize electricity usage might need to share...

Policy derivation methods for critic-only reinforcement learning in continuous spaces

This paper addresses the problem of deriving a policy from the value function in the context of critic-only reinforcement learning (RL) in continuous state and action spaces. With continuous-valued states, RL algorithms have to rely on a numerical approximator to represent the value function. Numerical approximation due to its nature virtually...