Posts by Collection

portfolio

publications

Self-Tuning PID Control via a Hybrid Actor-Critic-Based Neural Structure for Quadcopter Control

Published in 30th Annual International Conference of Iranian Society of Mechanical Engineers (ISME), 2022

Introduces a hybrid actor-critic neural structure for self-tuning PID control of quadcopters, improving dynamic response and robustness.

Recommended citation: I. Sharifi and A. Alasty. (2022). "Self-Tuning PID Control via a Hybrid Actor-Critic-Based Neural Structure for Quadcopter Control." ISME Conference, 2022.
Download Paper

Symbolic Imitation Learning: From Black-Box to Explainable Driving Policies

Published in Applied Sciences, 2025

Introduces a symbolic imitation learning approach using inductive logic programming to extract explainable behavioral rules from human demonstrations in highway driving.

Recommended citation: I. Sharifi, M. Yildirim, and S. Fallah. (2025). "Symbolic Imitation Learning: From Black-Box to Explainable Driving Policies." Applied Sciences.
Download Paper

Towards Safe Autonomous Highway Driving Policies using a Neuro-Symbolic Deep Reinforcement Learning Approach

Published in Transportation Research Record (TRR), 2025

A neuro-symbolic deep reinforcement learning framework integrating symbolic reasoning and neural control to ensure safety and interpretability in autonomous driving systems.

Recommended citation: I. Sharifi, M. Yildirim, and S. Fallah. (2025). "Towards Safe Autonomous Driving Policies using a Neuro-Symbolic Deep Reinforcement Learning Approach." Transportation Research Record, 2026.
Download Paper

Agricultural Spraying Drones: A Comprehensive Review

Published in Smart Agricultural Technology, 2025

A comprehensive review of agricultural spraying drones, covering drone architectures, control mechanisms, droplet deposition, and path planning techniques.

Recommended citation: A. Talaeizadeh, I. Sharifi, Sh. G. Samani, and A. Alasty. (2025). "Agricultural Spraying Drones: A Comprehensive Review." Smart Agricultural Technology, 2025.
Download Paper

Integrating Neurosymbolic AI in Advanced Air Mobility: A Comprehensive Survey

Published in International Joint Conference on Artificial Intelligence (IJCAI), 2025

A comprehensive survey of neurosymbolic artificial intelligence methods and their integration into Advanced Air Mobility (AAM) applications.

Recommended citation: K. Acharya, I. Sharifi, M. Lad, L. Sun, and H. H. Song. (2025). "Integrating Neurosymbolic AI in Advanced Air Mobility: A Comprehensive Survey." IJCAI, 2025.
Download Paper

A Survey of Security Challenges and Solutions for Advanced Air Mobility and eVTOL Aircraft

Published in AIAA SciTech Forum, 2026

Survey identifying and categorizing security vulnerabilities and mitigation strategies in Advanced Air Mobility (AAM) and eVTOL ecosystems.

Recommended citation: M. Ghazanfari, I. Sharifi, A. Taye, P. Wei, G. Biswas, B. Ward, X. Koutsoukos, M. Ghasemi, V. Gupta, A. Chen, A. Shirkhodaei, S. Hasan, I. Amundson, F. Fotiadis, U. Topcu, et al. (2026). "A Survey of Security Challenges and Solutions for Advanced Air Mobility and eVTOL Aircraft." AIAA SciTech Forum, 2026.
Download Paper

A Survey of Security Challenges and Solutions for UAS Traffic Management (UTM) and Small Unmanned Aerial Systems (sUAS)

Published in AIAA SciTech Forum, 2026

Comprehensive survey of cybersecurity challenges and solutions in UAS Traffic Management and small Unmanned Aerial Systems, with emphasis on communication, authentication, and operational vulnerabilities.

Recommended citation: I. Sharifi, M. Ghazanfari, A. Taye, P. Wei, G. Biswas, B. Ward, X. Koutsoukos, M. Ghasemi, V. Gupta, A. Chen, A. Shirkhodaei, S. Hasan, I. Amundson, F. Fotiadis, U. Topcu, et al. (2026). "A Survey of Security Challenges and Solutions for UAS Traffic Management (UTM) and small Unmanned Aerial Systems (sUAS)." AIAA SciTech Forum, 2026.
Download Paper

Reinforcement Learning Enhancement Using Vector Semantic Representation and Symbolic Reasoning for Human-Centered Autonomous Emergency Braking

Published in arXiv preprint, 2026

A neuro-symbolic pipeline that fuses semantic, spatial, and shape features with a Soft First-Order Logic reward — extracting symbolic predicates from segmentation maps — to improve the robustness and safety of camera-based deep reinforcement learning for autonomous emergency braking, evaluated in CARLA.

Recommended citation: V. Asodia, I. Sharifi, and S. Fallah. (2026). "Reinforcement Learning Enhancement Using Vector Semantic Representation and Symbolic Reasoning for Human-Centered Autonomous Emergency Braking." arXiv preprint arXiv:2602.05079.
Download Paper

Separation Assurance between Heterogeneous Fleets of Small Unmanned Aerial Systems via Multi-Agent Reinforcement Learning

Published in IEEE International Conference on Intelligent Transportation Systems (ITSC), 2026

A decentralized-centralized multi-agent reinforcement learning framework for ensuring separation assurance and fairness among heterogeneous sUAS fleets in shared airspace.

Recommended citation: I. Sharifi, H. T. Kim, M. H. Ahmed, M. Ghasemi, and P. Wei. (2026). "Separation Assurance between Heterogeneous Fleets of Small Unmanned Aerial Systems via Multi-Agent Reinforcement Learning." IEEE ITSC, 2026.
Download Paper

Fine-Tuning Large Language Models for Cooperative Tactical Deconfliction of Small Unmanned Aerial Systems

Published in Proceedings of the Computer Vision and Pattern Recognition (CVPR) Workshops, 2026

Explores supervised fine-tuning and Group Relative Policy Optimization (GRPO) to adapt large language models for centralized conflict resolution and safe separation assurance in cooperative multi-agent sUAS operations.

Recommended citation: I. Sharifi, A. Zongo, and P. Wei. (2026). "Fine-Tuning Large Language Models for Cooperative Tactical Deconfliction of Small Unmanned Aerial Systems." CVPR Workshops, 2026.
Download Paper

ANDRE: An Attention-based Neuro-symbolic Differentiable Rule Extractor for Inductive Logic Programming

Published in International Conference on Neurosymbolic Learning and Reasoning (NeSy), 2026

A neuro-symbolic differentiable Inductive Logic Programming framework that learns first-order logic rules over a continuous rule space using attention-based conjunction/disjunction operators, enabling accurate, stable, and interpretable rule extraction from noisy, probabilistic data.

Recommended citation: I. Sharifi, P. Wei, and S. Fallah. (2026). "ANDRE: An Attention-based Neuro-symbolic Differentiable Rule Extractor for Inductive Logic Programming." International Conference on Neurosymbolic Learning and Reasoning (NeSy), 2026.
Download Paper

talks

Poster Presentation: sUAS Separation Assurance with Mixed Aircraft Equipage and Heterogeneous Tactical Deconfliction Policies

Published: October 16, 2025

A multi-agent reinforcement learning framework for safe separation among heterogeneous small UAS fleets in shared airspace, presented at the NASA ULI program workshop.

Poster Presentation: Fine-Tuning Large Language Models for Cooperative Tactical Deconfliction of Small Unmanned Aerial Systems

Published: June 04, 2026

Fine-tuning large language models with supervised fine-tuning and GRPO for centralized conflict resolution and safe separation assurance in cooperative multi-agent drone operations.

Iman Sharifi