<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.4.1">Jekyll</generator><link href="https://gotvandhub.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="https://gotvandhub.github.io/" rel="alternate" type="text/html" /><updated>2026-11-08T16:58:56+00:00</updated><id>https://gotvandhub.github.io/feed.xml</id><title type="html">Gotvandhub</title><subtitle>Your Name&apos;s academic portfolio</subtitle><author><name>Yousef Emami</name><email>yousef.emami@gmail.com</email></author><entry><title type="html">The Unbiased Arbiter: Teaching Robots to Be Fair Moderators</title><link href="https://gotvandhub.github.io/posts/2026/11/arbiter/" rel="alternate" type="text/html" title="The Unbiased Arbiter: Teaching Robots to Be Fair Moderators" /><published>2026-11-04T00:00:00+00:00</published><updated>2026-11-04T00:00:00+00:00</updated><id>https://gotvandhub.github.io/posts/2026/11/arbiter</id><content type="html" xml:base="https://gotvandhub.github.io/posts/2026/11/arbiter/"><![CDATA[<p>Recently, I read “Superintelligence: Paths, Dangers, Strategies” by Nick Bostrom. It argues that superintelligence, if created, would be difficult to control and that it could take over the world to accomplish its goals. In this short article, I want to add my perspective to make superintelligence a collaborator for humans. In my view, we should train robots to become unbiased learners and moderators. My vision was rooted in the belief that just as genuine humans can become unbiased over time by embracing their true selves, robots could be taught to act as fair moderators by considering multiple perspectives before taking any action. I propose a multifaceted approach to achieve this, involving learning through conflict, embedding ethical guidelines, and transparency.</p>

<p>What do I mean by learning through conflict (LTC)? What motivates me? The basic idea of LTC is that when a human faces contradictory facts, they gradually become a judge and stand alone to make unbiased and independent decisions. We plan to enable an artificial intelligence (AI) agent to conduct LTC by training with contradictory data. First, the agent is inputted and trained with contradictory data comprising current data and a divergent perspective(s) of the data. Then, by adopting some strategies, early convergence to the data or its perspective(s) is prevented. The main advantage of this method is the unbiased output by learning the underlying features of both data and its perspective(s) and making the AI agent become a judge capable of independent decisions.</p>

<p>The objective of learning through conflict is developing ethical norms embedding ethical guidelines into AI systems, and making moderation as a middle value for them. Also, transparency in AI decision-making processes fosters trust and facilitates the identification and mitigation of biases.</p>]]></content><author><name>Yousef Emami</name><email>yousef.emami@gmail.com</email></author><category term="AI" /><category term="Robots" /><category term="Arbiter" /><summary type="html"><![CDATA[Recently, I read “Superintelligence: Paths, Dangers, Strategies” by Nick Bostrom. It argues that superintelligence, if created, would be difficult to control and that it could take over the world to accomplish its goals. In this short article, I want to add my perspective to make superintelligence a collaborator for humans. In my view, we should train robots to become unbiased learners and moderators. My vision was rooted in the belief that just as genuine humans can become unbiased over time by embracing their true selves, robots could be taught to act as fair moderators by considering multiple perspectives before taking any action. I propose a multifaceted approach to achieve this, involving learning through conflict, embedding ethical guidelines, and transparency.]]></summary></entry><entry><title type="html">Human-In-The-Loop Reinforcement Learning in Autonomous Vehicles</title><link href="https://gotvandhub.github.io/posts/2013/08/human/" rel="alternate" type="text/html" title="Human-In-The-Loop Reinforcement Learning in Autonomous Vehicles" /><published>2026-11-04T00:00:00+00:00</published><updated>2026-11-04T00:00:00+00:00</updated><id>https://gotvandhub.github.io/posts/2013/08/human</id><content type="html" xml:base="https://gotvandhub.github.io/posts/2013/08/human/"><![CDATA[<p>Human-In-The-Loop Reinforcement Learning (HITL-RL) in Autonomous Vehicles (AVs) can be realized through reward shaping, human-initiated actions, and interactive learning to guide the AVs in critical situations. This can improve the efficiency of tasks and the learning ability and performance of the AVs. In addition, humans and AVs can interact and cooperate to solve tasks. In this regard, it should be noted that long-term monitoring and guidance can be physically and mentally demanding for human participants in the HITL-RL. Further, current HITL-RL methods often require expert-level demonstrations to ensure data quality and optimize performance improvement. However, expert guidance is expensive for largescale real-world applications. The main benefit of HITL-RL for AVs is to promote public trust and enhance safety.</p>

<p>Real-time human-UAV interactions using wearable gloves and Machine Lear (ML) based methods showcase innovative technology’s potential to enhance mission planning and control. Human and Unmanned Aerial Vehicle (UAV) interactions can be described using meta-models in emergency response scenarios to highlight the importance of formalizing human intervention points to ensure successful mission execution. HITL-RL frameworks are versatile and can be adapted to various applications, from Autonomous Driving (AD) to underwater exploration and emergency response. This versatility underlines the broad applicability and potential impact of HITL approaches in enhancing AVs across different domains. Incorporating mechanisms for human behavior modeling reduces the burden on human operators, making the system more efficient and user-friendly.</p>

<p>Developing a framework that learns reward functions from human driving data can effectively manage the learning process for AD, ensuring the RL agent optimizes complex goals. This approach includes trajectory sampling, offline preference learning, and RL integration to model human decision-making and obtain reward functions. Incorporating human prior experience into the reward function and safety constraints helps guide AVs in complex scenarios, such as mandatory lane changes, and limits unsafe exploration during training. Enhancing HITL-RL methods by having humans supervise, intervene, and demonstrate during the learning process not only improves the safety of AVs but also optimizes traffic flow efficiency in mixed-traffic environments. Creating a value-based HITL-DRL algorithm to address high-level decision-making problems, such as multiobjective lane-change decisions, improves training efficiency and adaptability. This approach values human policy over the undertrained DRL policy, allowing the RL agent to mimic human behaviors more efficiently. Enhancing RL performance by using human-guided learning objectives, prioritized human experience replay, and human intervention-based reward shaping improves the navigation capabilities of UGVs.</p>]]></content><author><name>Yousef Emami</name><email>yousef.emami@gmail.com</email></author><category term="Human-in-the-loop" /><category term="RL" /><category term="ML" /><summary type="html"><![CDATA[Human-In-The-Loop Reinforcement Learning (HITL-RL) in Autonomous Vehicles (AVs) can be realized through reward shaping, human-initiated actions, and interactive learning to guide the AVs in critical situations. This can improve the efficiency of tasks and the learning ability and performance of the AVs. In addition, humans and AVs can interact and cooperate to solve tasks. In this regard, it should be noted that long-term monitoring and guidance can be physically and mentally demanding for human participants in the HITL-RL. Further, current HITL-RL methods often require expert-level demonstrations to ensure data quality and optimize performance improvement. However, expert guidance is expensive for largescale real-world applications. The main benefit of HITL-RL for AVs is to promote public trust and enhance safety.]]></summary></entry><entry><title type="html">LLM-Centric Agentic AI for UAV Swarms</title><link href="https://gotvandhub.github.io/posts/2026/11/llm-centric-agentic-ai-uav-swarms/" rel="alternate" type="text/html" title="LLM-Centric Agentic AI for UAV Swarms" /><published>2026-11-04T00:00:00+00:00</published><updated>2026-11-04T00:00:00+00:00</updated><id>https://gotvandhub.github.io/posts/2026/11/llm-centric-agentic-ai-uav-swarms</id><content type="html" xml:base="https://gotvandhub.github.io/posts/2026/11/llm-centric-agentic-ai-uav-swarms/"><![CDATA[<p>UAV swarms promise a lot for search and rescue, environmental monitoring, and precision agriculture, but real deployment is held back by limited situational awareness and intermittent connectivity. Standard LLMs don’t close that gap on their own: they lack temporal memory and native tool interaction, and they can’t generate verifiable, real-time control actions in safety-critical settings. Our paper, LLM-Centric Agentic AI for UAV Swarms, proposes LAUS, a system-level architecture that uses agentic AI to address this.</p>

<p>The central design choice is a strict separation of timescales. The inner loop (milliseconds, onboard, deterministic) handles flight control, reactive collision avoidance, and safety validation, and never invokes the LLM. The mid loop (sub-second to seconds) runs deterministic solvers for trajectory optimization, formation control, and resource allocation, exposed to the agent only as callable tools. The outer loop (seconds to minutes, edge-hosted) is the only tier where the LLM operates, selecting objectives, re-parameterizing solver calls, and scheduling sensing. LLM latency and connectivity outages therefore bound mission-level adaptation, not flight-critical control, which continues under the last validated set-points.</p>

<p>Around this core, LAUS has four main components. Context engineering is the system-level layer that decides what the model sees, remembers, and uses across steps, turning a stateless LLM into an agentic one by managing memory, retrieval, context assembly, and tool use. Tool calling serves every stage of the loop: perception (parsing LiDAR, thermal, and GNSS/IMU data into structured observations), reasoning (retrievers over knowledge bases such as 3GPP standards), and action (resource allocation, formation managers, weather predictors), with outputs gated through independent safety validators before reaching the flight controller.</p>

<p>Agentic memory stores experiences as atomic, interconnected notes with embeddings, rather than rigid predefined schemas. New experiences are encoded as notes, linked to similar past situations, and updated through semantic association, which helps agents keep making decisions during communication blackouts using locally cached, historically grounded context. Adaptation across missions happens without any parametric updates to the LLM: what changes over time is the assembled context, not the model weights.</p>

<p>Finally, a governance layer with human-in-the-loop oversight provides enforceable rules, audit logs for traceability, mandatory human approval for critical decisions, and a fail-safe kill switch, tracked through KPIs such as communication reliability, collision avoidance, energy efficiency, and mission success rate. One note on scope: the paper specifies memory, governance, and oversight as architectural mechanisms; they are not independently instantiated or empirically evaluated.</p>]]></content><author><name>Yousef Emami</name><email>yousef.emami@gmail.com</email></author><category term="LLM" /><category term="UAV Swarms" /><category term="Agentic AI" /><summary type="html"><![CDATA[UAV swarms promise a lot for search and rescue, environmental monitoring, and precision agriculture, but real deployment is held back by limited situational awareness and intermittent connectivity. Standard LLMs don’t close that gap on their own: they lack temporal memory and native tool interaction, and they can’t generate verifiable, real-time control actions in safety-critical settings. Our paper, LLM-Centric Agentic AI for UAV Swarms, proposes LAUS, a system-level architecture that uses agentic AI to address this.]]></summary></entry><entry><title type="html">When Machine Learning Meets UAVs Challenges: A Starter Guide</title><link href="https://gotvandhub.github.io/posts/2026/11/starter/" rel="alternate" type="text/html" title="When Machine Learning Meets UAVs Challenges: A Starter Guide" /><published>2026-11-04T00:00:00+00:00</published><updated>2026-11-04T00:00:00+00:00</updated><id>https://gotvandhub.github.io/posts/2026/11/starter</id><content type="html" xml:base="https://gotvandhub.github.io/posts/2026/11/starter/"><![CDATA[<p>Unmanned Aerial Vehicles (UAVs) are becoming indispensable in modern communication networks, serving as key components in 5G and anticipated to drive 6G networks. However, UAVs face intricate challenges that demand Machine Learning (ML) solutions. This article examines ML solutions tailored to address the challenges and aims to guide researchers and PhD students in advancing UAV technology and its integration with ML.</p>

<ol>
  <li>Interference Management</li>
</ol>

<p>ML, particularly deep reinforcement learning (DRL), is essential for addressing interference management in UAVs because traditional methods depend on detailed channel information, which results in high signaling overhead and impractical implementation. ML techniques can learn and adapt to interference patterns, enabling efficient management without the need for explicit channel information, thus reducing overhead and improving the feasibility of interference mitigation in dynamic UAV environments.</p>

<ol>
  <li>Energy Efficiency and Power Control</li>
</ol>

<p>We require ML solutions for energy efficiency and power control in UAVs because UAVs have limited energy resources, making the optimization of energy and power consumption crucial for the overall performance of the network. Various ML solutions have been proposed to exploit the network’s resources in a near-optimal manner, aiming to minimize energy consumption effectively.</p>

<ol>
  <li>Channel Modeling</li>
</ol>

<p>We require ML for channel modeling in UAVs because accurately predicting channel behavior in wireless networks is crucial, especially in highly mobile environments. Selecting appropriate channel models is essential for performance evaluation, network planning, and UAV trajectory design. Various ML solutions have been proposed for predicting channel behavior and UAV-received signal strength (RSS). ML techniques can assess wireless conditions effectively without requiring the extensive datasets typically needed for traditional channel modeling based on extensive measurement campaigns.</p>

<ol>
  <li>UAV Detection</li>
</ol>

<p>ML is essential for UAV detection because it enables accurate, real-time identification of UAV-specific RF signatures. ML adapts to new data, handles diverse environments, reduces false positives, efficiently processes large data sets, and integrates well with existing systems, ensuring a robust and reliable detection mechanism.</p>

<ol>
  <li>Collision Avoidance</li>
</ol>

<p>We require ML solutions for UAVs collision avoidance because avoiding collisions with both fixed and moving environmental obstacles, as well as between UAVs in a swarm, is crucial for safe and efficient operations. Vision-based approaches, such as deep learning, can assist UAVs in accurately perceiving their environment and navigating safely.</p>

<ol>
  <li>Spectrum Management</li>
</ol>

<p>ML is essential for UAV spectrum management because it enables accurate prediction of spectrum availability, dynamic adaptation of transmission parameters, efficient allocation of resources, interference mitigation, enhanced spectrum utilization, scalability, and reduced need for human intervention. This ensures optimal and reliable UAV communication.</p>

<ol>
  <li>Data Collection Scheduling</li>
</ol>

<p>ML is essential for data collection scheduling using UAVs because it enables dynamic adaptation to changing conditions, effectively manages large state and action spaces, and addresses the challenge of outdated sensor data. By utilizing adaptive algorithms such as DRL, UAVs can optimize data collection by continuously adjusting to real-time environmental changes, selecting the best sensors based on current conditions, and making informed decisions despite intermittent updates. This ensures efficient, reliable, and autonomous data collection operations, enhancing overall system performance.</p>

<ol>
  <li>Routing</li>
</ol>

<p>The UAV-enabled networks are a new communication paradigm that is characterized by high-speed mobility and frequent topology modification. In that sense, routing protocols are not able to fully support the highly dynamic nature of the flying ad hoc UAV communication networks. A promising approach that could improve this aspect is to adopt ML in routing protocols, leading to autonomous network coordination and distributed user scheduling and data routing through dynamic path formation.</p>

<ol>
  <li>Placement and Path Planning</li>
</ol>

<p>ML is essential for UAV placement and path planning due to the complex nature of designing UAV-based communication systems and the critical requirements in applications such as public safety and security. Determining the optimal horizontal and vertical placement of UAVs, as well as planning their trajectories to interact with ground or flying objects, requires sophisticated algorithms that can handle large-scale optimization problems and dynamic environmental factors. In a nutshell, ML is indispensable for UAV placement and path planning and facilitates complex optimization tasks, supports real-time adaptation to changing conditions, and enhances capabilities for detecting and responding to noncompliant UAVs in critical applications like public safety and security.</p>

<ol>
  <li>Localization</li>
</ol>

<p>ML is crucial for localization in UAV networks because while exact optimization techniques can accurately solve localization problems, they often fall short in providing real-time calculations required for dynamic applications. ML-based techniques offer a compelling alternative by leveraging their ability to learn from data and adapt in real time. This capability allows them to provide near-optimal localization solutions swiftly and practically, meeting the stringent timing requirements of UAV operations. By continuously learning from incoming sensor data and environmental changes, ML enables UAVs to localize themselves accurately and efficiently in dynamic and challenging environments, ensuring reliable performance in critical applications such as surveillance, disaster response, and autonomous navigation.</p>

<ol>
  <li>Security</li>
</ol>

<p>ML is crucial for UAV security because it offers adaptive and accurate solutions to combat GPS spoofing, jamming, intrusion detection, and other emerging threats. Its ability to analyze vast amounts of data in real time and adapt to evolving threat landscapes makes it indispensable for ensuring the safety, reliability, and integrity of UAV operations in both civilian and critical applications.</p>

<p>In conclusion, integrating ML solutions into UAV-enabled communication networks is crucial for overcoming the challenges discussed. These challenges underscore the need for continual innovation and research across various layers of the OSI reference model, from physical to application. Researchers and PhD students can utilize this comprehensive overview to delve deeper into these critical areas, driving forward advancements that will shape the future of UAV-enabled communication systems.</p>]]></content><author><name>Yousef Emami</name><email>yousef.emami@gmail.com</email></author><category term="ML" /><category term="UAVs" /><category term="Trajectory" /><summary type="html"><![CDATA[Unmanned Aerial Vehicles (UAVs) are becoming indispensable in modern communication networks, serving as key components in 5G and anticipated to drive 6G networks. However, UAVs face intricate challenges that demand Machine Learning (ML) solutions. This article examines ML solutions tailored to address the challenges and aims to guide researchers and PhD students in advancing UAV technology and its integration with ML.]]></summary></entry></feed>