Equivariant Ensembles and Regularization for Reinforcement Learning in Map-based Path Planning

no code implementations • 19 Mar 2024 • Mirco Theile, Hongpeng Cao, Marco Caccamo, Alberto L. Sangiovanni-Vincentelli

In reinforcement learning (RL), exploiting environmental symmetries can significantly enhance efficiency, robustness, and performance.

Inductive Bias Reinforcement Learning (RL)

Learning to Recharge: UAV Coverage Path Planning through Deep Reinforcement Learning

1 code implementation • 6 Sep 2023 • Mirco Theile, Harald Bayerlein, Marco Caccamo, Alberto L. Sangiovanni-Vincentelli

Coverage path planning (CPP) is a critical problem in robotics, where the goal is to find an efficient path that covers every point in an area of interest.

reinforcement-learning

Edge Generation Scheduling for DAG Tasks Using Deep Reinforcement Learning

1 code implementation • 28 Aug 2023 • Binqi Sun, Mirco Theile, Ziyuan Qin, Daniele Bernardini, Debayan Roy, Andrea Bastoni, Marco Caccamo

Using this schedulability test, we propose a new DAG scheduling framework (edge generation scheduling -- EGS) that attempts to minimize the DAG width by iteratively generating edges while guaranteeing the deadline constraint.

reinforcement-learning Scheduling

Learning to Generate All Feasible Actions

no code implementations • 26 Jan 2023 • Mirco Theile, Daniele Bernardini, Raphael Trumpp, Cristina Piazza, Marco Caccamo, Alberto L. Sangiovanni-Vincentelli

Several machine learning (ML) applications are characterized by searching for an optimal solution to a complex task.

Cloud-Edge Training Architecture for Sim-to-Real Deep Reinforcement Learning

no code implementations • 4 Mar 2022 • Hongpeng Cao, Mirco Theile, Federico G. Wyrwal, Marco Caccamo

To overcome the reality gap, our architecture exploits sim-to-real transfer strategies to continue the training of simulation-pretrained agents on a physical system.

Domain Adaptation reinforcement-learning +1

Multi-UAV Path Planning for Wireless Data Harvesting with Deep Reinforcement Learning

1 code implementation • 23 Oct 2020 • Harald Bayerlein, Mirco Theile, Marco Caccamo, David Gesbert

Harvesting data from distributed Internet of Things (IoT) devices with multiple autonomous unmanned aerial vehicles (UAVs) is a challenging problem requiring flexible path planning methods.

Collision Avoidance Multi-agent Reinforcement Learning +2

UAV Path Planning using Global and Local Map Information with Deep Reinforcement Learning

2 code implementations • 14 Oct 2020 • Mirco Theile, Harald Bayerlein, Richard Nai, David Gesbert, Marco Caccamo

Path planning methods for autonomous unmanned aerial vehicles (UAVs) are typically designed for one specific type of mission.

reinforcement-learning Reinforcement Learning (RL)

UAV Path Planning for Wireless Data Harvesting: A Deep Reinforcement Learning Approach

3 code implementations • 1 Jul 2020 • Harald Bayerlein, Mirco Theile, Marco Caccamo, David Gesbert

Autonomous deployment of unmanned aerial vehicles (UAVs) supporting next-generation communication networks requires efficient trajectory planning methods.

reinforcement-learning Reinforcement Learning (RL) +1

UAV Coverage Path Planning under Varying Power Constraints using Deep Reinforcement Learning

2 code implementations • 5 Mar 2020 • Mirco Theile, Harald Bayerlein, Richard Nai, David Gesbert, Marco Caccamo

Coverage path planning (CPP) is the task of designing a trajectory that enables a mobile agent to travel over every point of an area of interest.

Robotics Systems and Control Systems and Control