# Paolo Baldini

> Paolo Baldini is a researcher at the University of Bologna working on artificial intelligence, adaptive robotics and multi-agent / collective systems, with publications on online adaptation of robots, Boolean networks, reservoir computing, nanowire networks and large language models.

Paolo Baldini works at the [University of Bologna](https://www.unibo.it/en/homepage), where he conducts research and teaching activities.
He is currently *adjunct professor* for the course of [Software Design and Development](https://www.unibo.it/en/study/course-units-transferable-skills-moocs/course-unit-catalogue/course-unit/2026/498808).
Furthermore, he is involved in the [FIS3 Starting Grant "Foundations for Macro-programming-based Software Engineering" (FoMaSE)](https://fis3-fomase.github.io), under the supervision of Ph.D. Roberto Casadei.

Paolo Baldini is involved in research in the field of *Artificial Intelligence*, *Adaptive Robotics*, and *Multi-Agent / Collective Systems*. His aims range from enabling robots to autonomously face unpredictable situations that might occur during operation, to better predict the dynamics of multi-agent systems.
Other areas of interest concern *Complex Systems*, *Unconventional Computing*, *Large Language Models*, and *Gene Regulatory Networks*.

Paolo Baldini is author and co-author of several research articles, and has been reviewer and program committee member for several international conferences.
As part of his teaching activities, he contributed to the courses of *Computer Architectures* and *Algorithms and Data Structures*; furthermore, he taught robot programming at several high-schools for the university orientation project "[In the role of a robot: how to program the behavior of a robot](https://site.unibo.it/scuola-universita/it/cosa-possiamo-fare-insieme-/catalogo-multicampus-iniziative-pcto-pls-pot/nei-panni-di-un-robot-come-programmare-il-comportamento-di-un-robot)".
He is currently supervising a master's thesis and has been co-supervisor of four other master's theses.

Website: https://p-baldini.github.io/
LinkedIn: https://www.linkedin.com/in/baldini-paolo
GitHub: https://github.com/p-baldini
X / Twitter: https://x.com/pbmandrab
Profile: https://orcid.org/0000-0002-3625-5414
Profile: https://www.semanticscholar.org/author/2095684681
Profile: https://openalex.org/A5044031617
Profile: https://scholar.google.com/citations?user=vUmVyzMAAAAJ
Profile: https://dblp.org/pid/86/6789

This summary, with the full text of the publications that provide it, is also available as a single file: https://p-baldini.github.io/llms-full.txt

## Journal articles

- [Unraveling creativity through variability: A comparison of LLMs and humans in an educational Q&A scenario](https://p-baldini.github.io/publications/2026-tkl/): Michele Braccini, Gianluca Aguzzi, Paolo Baldini. Technology, Knowledge and Learning, February 2026. DOI: 10.1007/s10758-026-09956-9. PDF: https://p-baldini.github.io/publications/2026-tkl/2026-tkl.pdf. Abstract: Large Language Models (LLMs) have demonstrated a remarkable capacity for generating human-quality text across diverse applications. In light of this, the use of these models is becoming increasingly widespread. Education is no exception and so literature is starting to explore the potential of LLMs in educational contexts: ranging from teaching assistant chatbot to personalized tutoring. Although the versatility of these models is evident, the critical question arises whether they can truly replicate the creativity and nuances of human languages, giving students the variability and richness of expression necessary for effective learning, and for nurturing critical and divergent thinking. This is particularly important in the pre-school or primary education context, where novel and insightful content is crucial for effective engagement. Framing this problem from an abstract linguistic perspective, it translates into wondering how effectively the “creativity” of LLM-generated text can be quantified and how it compares to human one. So, recognizing the inherent ambiguity in defining creativity this paper investigates the variability of LLM outputs as a potential proxy, and compares it with observed text variability in a large human-authored dataset. Specifically focusing on early childhood education (ECE), we compare the semantic and syntactic diversity of responses from LLMs and humans prompted to answer like 5-year-olds. Our empirical results show LLMs have lower variability than humans, particularly on repeated questions, though the gap narrows for different questions. This limitation raises concerns for educational suitability, where diverse explanations of concepts are crucial, suggesting a need to enhance LLM expressive range and diversity.
- [Fault Recovery through Online Adaptation of Boolean Network robots](https://p-baldini.github.io/publications/2025-sensors/): Paolo Baldini, Michele Braccini, Andrea Roli. Sensors, September 2025. DOI: 10.3390/s25185849. PDF: https://p-baldini.github.io/publications/2025-sensors/2025-sensors.pdf. Abstract: Being able to recover from faults is a desired capability in robotics. This requires identifying ineffective behaviors and making some changes so as to display the desired one. In this work, we consider the problem of adjusting the controller of a robot so as to produce the desired behavior. Instead of considering complex and ad-hoc modifications, we leverage the automatic discovery of suitable solutions by means of online adaptation, a mechanism for the modification of the robot control strategy in runtime. Specifically, we use a performance function to identify ineffective behaviors and drive the controller design to an effective one. We also discuss the technical requirements for this procedure to succeed. The results suggest that online adaptation is suitable for the automatic recovery of functions after the occurrence of damages. Additionally, we show that adapting an existing controller to overcome a fault is faster than searching for a new controller from scratch.
- [Cell–Cell interactions: how coupled Boolean Networks tend to criticality](https://p-baldini.github.io/publications/2025-artificial_life/): Michele Braccini, Paolo Baldini, Andrea Roli. Artificial Life, February 2025. DOI: 10.1162/artl_a_00444. PDF: https://p-baldini.github.io/publications/2025-artificial_life/2024-artificial_life.pdf. Abstract: Biological cells are usually operating in conditions characterized by intercellular signaling and interaction, which are supposed to strongly influence individual cell dynamics. In this work, we study the dynamics of interacting random Boolean networks, focusing on attractor properties and response to perturbations. We observe that the properties of isolated critical Boolean networks are substantially maintained also in interaction settings, while interactions bias the dynamics of chaotic and ordered networks toward that of critical cells. The increase in attractors observed in multicellular scenarios, compared to single cells, allows us to hypothesize that biological processes, such as ontogeny and cell differentiation, leverage interactions to modulate individual and collective cell responses.
- [On the performance of Online Adaptation of robots controlled by Nanowire Networks](https://p-baldini.github.io/publications/2023-ieee_access/): Paolo Baldini, Andrea Roli, Michele Braccini. IEEE Access, December 2023. DOI: 10.1109/ACCESS.2023.3345224. PDF: https://p-baldini.github.io/publications/2023-ieee_access/2023-ieee_access.pdf. Abstract: In order to complete useful tasks in complex and changing contexts, robots need to be able to adapt their behavior or actions. This requires the ability to find effective approaches to the situation at hand, and to maintain them until they keep being effective. The final goal can be summarized by the will of maintaining a high performance during the whole life of the robot, regardless of the changes that may intervene during its activity. In this work, we evaluate by the point of view of the resulting life-long performance two methodologies for the adaptation of robots controlled by immutable network-based control systems: the Nanowire Networks. We demonstrate that modifying the best found solution leads to constant improvement and to overall better cumulative performance. Complementarily, we show that a less constrained approach simplifies the exploration of different behaviors but reduces life-long performance. Finally, we confirm previous results suggesting the potential of using this novel neuromorphic device (i.e., the Nanowire Network) for the control of robots.

## Conference and workshop papers

- [Macro-to-micro behavioural mappings in distributed systems: a characterisation on event structures](https://p-baldini.github.io/publications/2026-sasso/): Roberto Casadei, Paolo Baldini, Niccolò Castronuovo. Fourth International Workshop on Sustainability and Scalability of Self-Organisation (SaSSO), Cesena, Italy, July 2026. PDF: https://p-baldini.github.io/publications/2026-sasso/2026-sasso.pdf. Abstract: Engineering the self-organising behaviour of artificial collective systems, such as robot swarms, is challenging. Effective abstractions, provided by programming paradigms tailored to the purpose, can simplify the task of designers by promoting conceptual and practical tools supporting reasoning and implementation. Macro-programming has emerged as an umbrella term for paradigms (like aggregate and choreographic programming) aimed at specifying collective behaviours by a global perspective, largely abstracting micro-level details. In this work, we characterise the macro-programming setting by leveraging the formal framework of event structures, which represents collective computations as causal graphs of interactions among computation events. Specifically, we identify different types of macro-to-micro mechanisms, namely ways of mapping macroscopic specifications into local policies, allowing to conceptualise notable paradigms like aggregate and choreographic programming as specific instances of macro-programming.
- [On the LLM Robustness in a Simulated Conversational XR Scenario: A Preliminary Semantic Analysis](https://p-baldini.github.io/publications/2026-huarl/): Michele Braccini, Gianluca Aguzzi, Paolo Baldini, Andrea Roli. ICIAP 2025 — 23rd International Conference on Image Analysis and Processing / HUARL 2025 — The 1st International Workshop on Humans in the eXtended Artificial Intelligence Loop, Rome, Italy, January 2026. DOI: 10.1007/978-3-032-11317-7_51. Abstract: Generative AI and eXtended Reality (XR) have introduced to computer applications not only new challenges, but also the need to reconsider from a new perspective classical problems to which hardware and software systems may be subject to and their implications. For example, errors or disturbances in the communication channel between the physical and virtual worlds can lead to systems malfunctions and unintended output, on the one hand, but can also have implications on the creativity of the XR system, on the other hand. Indeed, in systems where Large Language Models (LLMs) mediate or generate responses, these processes can trigger novel and unexpected outputs, thereby representing affordances for creative interactions. In this paper, we explore this scenario by starting analyzing the robustness of LLMs to perturbations—in the form of scrambled character blocks of prompts—thus simulating a typical XR chat scenario where an avatar is driven by an LLM. As a preliminary study, we analyze the response behavior of two small LLMs using a limited sample size of 30 questions.
- [Impact of LLM—assisted coding in creativity and robustness of robot controllers](https://p-baldini.github.io/publications/2025-haic/): Paolo Baldini, Michele Braccini, Andrea Roli. HAIC 2025 — First International Workshop on Human-AI Collaborative Systems, Bologna, Italy, October 2025. PDF: https://p-baldini.github.io/publications/2025-haic/2025-haic.pdf. Abstract: The use of Large Language Models (LLMs) in work environments has recently started to gain attention in the research community, with many works reporting increased productivity and others reporting homogenization of the output products. Nevertheless, their use as coding assistants in robotics has been mostly overlooked, especially from the point of view of their effects compared to not-assisted programming. We claim that peculiar characteristics of robotics programming deserve special attentions, such as the robustness of the produced solution. Here we analyze the effects of using LLMs as coding assistants in robotics. We analyze their effects on the performance, in a pseudo-reality gap, and on the similarity of the produced controllers. We also briefly discuss the feedback of some participants of the experiment. The results suggest that the codes produced with the assistance of LLMs are less robust to unseen conditions, and overall more homogeneous. Additionally, we report a shorter development time when using LLMs, but a poorer coding experience.
- [Emergent dynamics in heterogeneous life-like Cellular Automata](https://p-baldini.github.io/publications/2024-wivace_b/): Aarati Shrestha, Felix Reimers, Sanyam Jain, Paolo Baldini, Michele Braccini, Andrea Roli, Stefano Nichele. XVIII International Workshop on Artificial Life and Evolutionary Computation (WIVACE), Namur, Belgium, July 2025. DOI: 10.1007/978-3-031-93631-9_1. PDF: https://p-baldini.github.io/publications/2024-wivace_b/2024-wivace_b.pdf. Abstract: The Game of Life (GoL), one well known 2D cellular automaton, does not typically ensure interesting long-term phenotypic dynamics. Therefore, while being Turing complete, GoL cannot be said to be open-ended. In this work, we extend GoL with the opportunity for local mutations, thus enabling a heterogeneous life-like cellular automaton guided by an evolutionary inner loop. Additionally, we introduce the concept of cell ageing to ensure that cell aliveness (activated by inheritance with variation, and controlled by ageing) and actual cell computation (governed by life-like rules on local neighborhoods) are kept conceptually separated. We conduct an experimental campaign to identify suitable parameters that produce long-term phenotypic dynamics and favor genotypic innovations.
- [Cooperation in evolved modular soft robots](https://p-baldini.github.io/publications/2024-wivace_a/): Paolo Baldini, Michele Braccini, Filippo Benvenuti, Eric Medvet, Andrea Roli, Francesco Rusin. XVIII International Workshop on Artificial Life and Evolutionary Computation (WIVACE), Namur, Belgium, July 2025. DOI: 10.1007/978-3-031-93631-9_13. PDF: https://p-baldini.github.io/publications/2024-wivace_a/2024-wivace_a.pdf. Abstract: Self-assembly is the ability of natural beings to aggregate in a suitable structure in order to tackle complex tasks that could not be solved by single individuals. Robotics tries to replicate this ability to solve complex problems using artificial embodied physical agents. Nevertheless, the potential of self-assembly in robotics has still to be fully exploited. Here we propose self-assembly with voxel-based soft robots, a type of modular soft robots. The goal is to evolve the controller of the robots, inducing the formation of a single structure to move beyond a hole in the terrain that could not be crossed by individual robots. We perform experiments in simulation in increasingly complex arenas. The results show that evolved robots are capable of forming a connected chain and overall succeed in their mission. We also discuss limitations and improvements of the proposed approach.
- [An investigation of graceful degradation in Boolean Network robots subject to Online Adaptation](https://p-baldini.github.io/publications/2024-wivace/): Michele Braccini, Paolo Baldini, Andrea Roli. XVII International Workshop on Artificial Life and Evolutionary Computation (WIVACE), Venice, Italy, March 2024. DOI: 10.1007/978-3-031-57430-6_16. PDF: https://p-baldini.github.io/publications/2024-wivace/2023-wivace.pdf. Abstract: The ability to resist to faults is a desired property in robotic systems. However, it is also hard to obtain, having to modify the behavior to face breakdowns. In this work we investigate the robustness against sensor faults in robots controlled by Boolean networks. These robots are subject to online adaptation—i.e., they adapt some structural properties while they actually act—for improving their performance at a simple task, namely phototaxis. We study their performance variation according to the number of faulty light sensors. The outcome is that Boolean network robots exhibit graceful degradation, as the performance decreases gently with the number of faulty sensors. We also observed that a moderate number of faulty sensors—especially if located contiguously—not only produces a negligible performance decrease, but it can be sometimes even beneficial. We argue that online adaptation is a key concept to achieve fault tolerance, allowing the robot to exploit the redundancy of sensor signals and properly tune the dynamics of the same Boolean network depending on the specific working sensor configuration.
- [Online Adaptation of robots controlled by Nanowire Networks: a preliminary study](https://p-baldini.github.io/publications/2022-wivace/): Paolo Baldini, Michele Braccini, Andrea Roli. XVI International Workshop on Artificial Life and Evolutionary Computation (WIVACE), Gaeta, Italy, April 2023. DOI: 10.1007/978-3-031-31183-3_14. PDF: https://p-baldini.github.io/publications/2022-wivace/2022-wivace.pdf. Abstract: The ability to adapt to changes and unexpected situations is a commonly acknowledged hallmark of autonomy and intelligence. In this work we take inspiration from biology for the definition of a robot able to continuously adapt to changes. Specifically, we define its control structure and the mechanism used to perform the adaptation. The former is based on the Reservoir Computing framework, on which the latter acts. The result is the design of an Online Adaptive Reservoir Computing system based on a novel memristive reservoir: the Nanowire Network. Finally, the robot is tested on three different tasks taking place in different arenas. The results are then discussed and compared with a baseline algorithm.

## Preprints

- [Reservoir Computing in robotics: a review](https://p-baldini.github.io/publications/2022-arxiv/): Paolo Baldini. arXiv, June 2022. DOI: 10.48550/arXiv.2206.11222. PDF: https://p-baldini.github.io/publications/2022-arxiv/2022-arxiv.pdf. Abstract: Reservoir Computing is a relatively new framework created to allow the usage of powerful but complex systems as computational mediums. The basic approach consists in training only a readout layer, exploiting the innate separation and transformation provided by the previous, untrained system. This approach has shown to possess great computational capabilities and is successfully used to achieve many tasks. This review aims to represent the current ‘state-of-the-art’ of the usage of Reservoir Computing techniques in the robotic field. An introductory description of the framework and its implementations is initially given. Subsequently, a summary of interesting applications, approaches, and solutions is presented and discussed. Considerations, ideas and possible future developments are proposed in the explanation.

## Theses (authored and supervised)

- [Instinct and reflex inspired safety mechanisms for adaptive robots](https://p-baldini.github.io/publications/2026-thesis_b/): Eleonora Bertoni. Supervisor: Andrea Roli, Co-Supervisor: Paolo Baldini, Michele Braccini. Università di Bologna — Alma Mater Studiorum, July 2026. PDF: https://p-baldini.github.io/publications/2026-thesis_b/2026-thesis_b.pdf. Abstract: Adaptive robotics aims at building fully autonomous robots that are able to overcome unexpected changes while operating in, possibly unpredictable, environments. In this context, robots can adapt relying on learning capabilities or, in general, any internal change that leads to a modified behavior. However, during these processes robots are more vulnerable and more likely to be exposed to damage. Although strategies for damage recovery are possible, preventing damage before it occurs is preferable and of fundamental importance. In this thesis, we take inspiration from the protective role of instincts and reflexes in nature to provide adaptive robots with a module of innate safety mechanisms. We design a set of instincts and reflexes to protect the robots from different sources of damage. We provide an implementation of these mechanisms and test it on a robot subject to online adaptation. The results suggest that this integration can improve the resilience and autonomy of robots.
- [Online adaptation of autonomous robots](https://p-baldini.github.io/publications/2026-thesis_a/): Paolo Baldini. Supervisor: Andrea Roli. Università di Bologna — Alma Mater Studiorum, March 2026. PDF: https://p-baldini.github.io/publications/2026-thesis_a/2026-thesis_a.pdf. Abstract: Robots diffusion is constantly increasing, yet their applications remain largely confined to industries and other highly controlled environments; the cause is their inability to address unexpected situations that can occur in the open world. We argue that endowing robots with the ability to adjust their own behavior can improve their operational efficacy in uncontrolled environments. In this dissertation, we investigate factors enabling the runtime modification of robot behaviors. The proposed operative approach consists in the systematic investigation and resolution of increasingly complex conditions. During the exploration, we employ robots with minimal computational capabilities, so as to investigate the factors enabling adaptation without them being obscured by the complexity of the control system employed; this also permits providing adaptive capabilities to simple robots, which might not be able to operate complex control software. The principal contributions of this dissertation include the identification of factors essential for adapting systems, among which stand out the need for self-evaluation, re-evaluation, and time- and intensity-modulated adaptation. Additionally, we highlight the importance of well-devised evaluation criteria and of their inherent characteristics, mainly the evaluation time and whether they tend to induce the emergence of a specific behavior or permit a broad spectrum of outcomes. We propose adaptive mechanisms capable of operating on various robot control architectures, including one designed offline by an automatic system with the explicit intent of optimizing its runtime adaptive capabilities. Finally, we present, define, and propose solutions to the problem of “situationality”, a condition in which external factors determine the performance of the robot regardless of its capabilities. Overall, this dissertation aims to contribute to the literature on online adaptation in robotics, shedding light on some frequently underestimated issues. We expect that the proposed solutions and considerations will support the future advancement of this discipline.
- [Mechanisms for Online Adaptation in robots: a comparative study](https://p-baldini.github.io/publications/2025-thesis/): Benedetta Pacilli. Supervisor: Andrea Roli, Co-Supervisor: Paolo Baldini. Università di Bologna — Alma Mater Studiorum, July 2025. PDF: https://p-baldini.github.io/publications/2025-thesis/2025-thesis.pdf. Abstract: Robots deployed in dynamic environments must be able to adapt autonomously to changing conditions and perturbations. This thesis examines online adaptation strategies for minimally cognitive robotic agents, with a focus on their ability to achieve and sustain high performance. We explore a range of adaptive controllers, including architectures inspired by Braitenberg vehicles and Artificial Neural Network-based strategies, from simple feed-forward topologies to recurrent networks with internal memory, each tested in navigation with a collision avoidance task. Our experimental results compare the performance of various mechanisms and adaptation policies, highlighting the trade-offs between reactivity, memory, and robustness in different online adaptation settings.
- [Evolution in materia: esperimenti con le Nanowire Networks](https://p-baldini.github.io/publications/2024-thesis_b/): Noemi Valentini. Supervisor: Andrea Roli, Co-Supervisor: Eric Medvet, Michele Braccini, Paolo Baldini. Università di Bologna — Alma Mater Studiorum, July 2024. PDF: https://p-baldini.github.io/publications/2024-thesis_b/2024-thesis_b.pdf. Abstract: In questo lavoro si esplorano le possibilità dell’applicazione di tecniche di computazione evolutiva per realizzare controlli e configurazioni opportune di dispositivi fisici costituiti da Nanowire Network (NN), reti di nanofili con dinamiche simili alla plasticità sinaptica dei neuroni. Queste tecniche si pongono l’obiettivo di superare i limiti della computazione classica e avvicinarsi a forme di computazione analogica e fisica (morphological computation). Gli esperimenti condotti hanno dimostrato che le NN possono essere configurate per realizzare operazioni logiche mediante l’Evolution in Materia (EIM). L’EIM utilizza la computazione evolutiva per manipolare materiali fisici al fine di ottenere comportamenti desiderati. L’obiettivo è far evolvere prodotti fisici già esistenti piuttosto che crearne di nuovi, consentendo di trovare soluzioni innovative a problemi complessi senza conoscere a priori le caratteristiche del materiale. L’architettura delle NN è diversa per ogni nuova fabbricazione; ciò comporta che i metodi di progettazione tradizionali non sono applicabili e risulta invece cruciale l’uso di approcci evolutivi. In questo studio è stata sperimentata l’EIM con il compito di trovare le configurazioni adatte alla realizzazione di porte logiche sulle NN. Uno dei punti di attenzione di queste nuove tecnologie riguarda riproducibilità e robustezza delle soluzioni prodotte; pertanto, è stato necessario porre una particolare attenzione nell’analisi delle soluzioni trovate riguardo questi due aspetti.
- [Self-assembly in voxel-based robots](https://p-baldini.github.io/publications/2024-thesis_a/): Filippo Benvenuti. Supervisor: Andrea Roli, Co-Supervisor: Eric Medvet, Michele Braccini, Paolo Baldini. Università di Bologna — Alma Mater Studiorum, March 2024. PDF: https://p-baldini.github.io/publications/2024-thesis_a/2024-thesis_a.pdf. Abstract: Self-assembly is a process whereby individual robotic units, without centralized control, autonomously organize into desired structures or patterns to achieve complex tasks. Its significance in the literature appears to be awaiting further exploration, as it has not quite reached its pinnacle or received the level of formal recognition it deserves. One aim of this master thesis is to provide a comprehensive understanding of self-assembly mechanisms, beginning with a detailed examination of its definitions, properties, and historical advancements, as outlined in related literature. Additionally, this thesis aims to advance the field through experimental research for further validating the feasibility of self-assembly mechanisms. Through rigorous analysis of the experimental outcomes, we aim to offer valuable insights that can enrich the existing body of knowledge in this area. Furthermore, we aim to lay the groundwork for potential future developments, thereby fostering continued exploration and expansion of the research started in this thesis.
- [Online adaptation of robots controlled by nanowire networks](https://p-baldini.github.io/publications/2022-thesis/): Paolo Baldini. Supervisor: Andrea Roli, Co-Supervisor: Michele Braccini. Università di Bologna — Alma Mater Studiorum, March 2022. PDF: https://p-baldini.github.io/publications/2022-thesis/2022-thesis.pdf. Abstract: The use of current computational systems is sometimes problematic in robotics, as their cost and power consumption might be too high to find space in low-cost robots. Additionally, their computational capabilities are notable, but not always suited for real-time operations. These limits drastically reduce the amount of applications that can be envisioned. A basic example is the use of micro and nanobots, that even add the issue of size. The need to find different technologies is therefore strong. Here, one alternative computational system is proposed and used: Nanowire Networks. These are novel types of electrical circuits, able to show dynamical properties. Their low-cost and consumption, in addition to their high computational capabilities, make them a perfect candidate for robotic applications. Here, this possibility is assessed, evaluating their use as robots controllers. This research begins with preliminary studies on their behaviour, and considerations about their use. The initial analysis is then used to define an online, adaptive learning approach, allowing the robot to exploit the network to adapt to different tasks and environments. The tested capabilities are: a simple collision avoidance, with fault tolerance considerations; a fast, reactive behaviour to avoid illegal areas; a memory aware behaviour, that can navigate a maze according to an initial stimulus. The results support the promising capabilities of the robotic controller. Additionally, the power of the online adaptation is clearly shown. Therefore, this thesis paves the way for a new type of computation in the robotic area, allowing plastic, fault-tolerant, cheap and efficient systems to be developed.
- [Progettazione e sviluppo di un modulo sensoristico e di comunicazione per la piattaforma robotica AlpaBot](https://p-baldini.github.io/publications/2019-thesis/): Paolo Baldini. Supervisor: Andrea Roli. Università di Bologna — Alma Mater Studiorum, October 2019. PDF: https://p-baldini.github.io/publications/2019-thesis/2019-thesis.pdf. Abstract: The use of current computational systems is sometimes problematic in robotics, as their cost and power consumption might be too high to find space in low-cost robots. Additionally, their computational capabilities are notable, but not always suited for real-time operations. These limits drastically reduce the amount of applications that can be envisioned. A basic example is the use of micro and nanobots, that even add the issue of size. The need to find different technologies is therefore strong. Here, one alternative computational system is proposed and used: Nanowire Networks. These are novel types of electrical circuits, able to show dynamical properties. Their low-cost and consumption, in addition to their high computational capabilities, make them a perfect candidate for robotic applications. Here, this possibility is assessed, evaluating their use as robots controllers. This research begins with preliminary studies on their behaviour, and considerations about their use. The initial analysis is then used to define an online, adaptive learning approach, allowing the robot to exploit the network to adapt to different tasks and environments. The tested capabilities are: a simple collision avoidance, with fault tolerance considerations; a fast, reactive behaviour to avoid illegal areas; a memory aware behaviour, that can navigate a maze according to an initial stimulus. The results support the promising capabilities of the robotic controller. Additionally, the power of the online adaptation is clearly shown. Therefore, this thesis paves the way for a new type of computation in the robotic area, allowing plastic, fault-tolerant, cheap and efficient systems to be developed.

## Posters

- [Adattamento Online di sistemi robotici](https://p-baldini.github.io/publications/2025-poster/): Paolo Baldini. Università di Bologna — Alma Mater Studiorum, September 2025. DOI: 10.6084/m9.figshare.30217519. PDF: https://p-baldini.github.io/publications/2025-poster/2025-poster.pdf. Abstract: Poster presenting the research topic of Paolo Baldini, exhibited at the “Giornata del Dottorando del Campus di Cesena, Università di Bologna” (transl., Cesena Campus Doctoral Student Day, University of Bologna). Event link: https://eventi.unibo.it/orientacarriera-campuscesena.

## Software contributions

- [AutoMoDe Lahmacun](https://p-baldini.github.io/contributions/automode/): An spin-off of the tool for the design of adaptable control software (IRIDIA Laboratory, Bruxelles, Belgium).
- [ARGoS3 simulator](https://p-baldini.github.io/contributions/argos3/): The Anaconda package of the famous swarm-robotics simulator (Anaconda.org).