In the last decades, industrial automation has seen an increasing employment of advanced robotic systems, such as mobile manipulators, to improve the efficiency, precision and safety of production processes. However, one of the most significant challenges for an effective usage of mobile manipulators lies in the optimal placement of their base, which is crucial in order to ensure high performance in terms of manipulability and execution time. This thesis proposes a pre-deployment optimization strategy for a robotic workstation equipped with a six-degrees-of-freedom manipulator mounted on a linear guide, used for packing operations of objects belonging to different categories. The output of the algorithm will serve as input for the real system, optimizing its performance from the start. The goal is to define an optimal sequence of pick-and-place operations, balancing manipulability and cycle time. Using the Particle Swarm Optimization (PSO) algorithm, the optimal base position associated to each operation is determined based on the mean manipulability measure during the execution. Additionally, by integrating the heuristic Best Fit algorithm, the placement of objects in boxes is optimized with the aim of minimizing unused space, while accounting for different categories. The application of the algorithm to an industrial use case, performed making use of the Tecnomatix Process Simulate environment, demonstrated that the PSO configuration with 20 particles and 20 iterations led to a 12.18% improvement in terms of manipulability compared to a "fixed base" strategy, with a 6.37% increase in cycle time, justified by the gain in reliability. This approach thus proves to be effective for real industrial contexts, optimizing space, resources and robotic performance.
Negli ultimi decenni l'automazione industriale ha visto un crescente impiego di sistemi robotici avanzati, come i manipolatori mobili, utilizzati per migliorare efficienza, precisione e sicurezza dei processi produttivi. Tuttavia, una delle più grandi criticità per un efficiente utilizzo dei manipolatori mobili riguarda il posizionamento ottimale della base, cruciale per garantire prestazioni elevate in termini di manipolabilità e tempi di esecuzione. Questa tesi propone un’ottimizzazione preliminare ("pre-deployment") per una stazione robotica dotata di un manipolatore a sei gradi di libertà su guida lineare, impiegata per operazioni di imballaggio di oggetti appartenenti a diverse categorie. L’output dell’algoritmo verrà pertanto utilizzato come input per il sistema reale, ottimizzandone le prestazioni sin dall’avvio. L’obiettivo è definire una sequenza ottimale di operazioni di "pick-and-place", bilanciando manipolabilità e tempo di ciclo. Mediante l'utilizzo dell'algoritmo Particle Swarm Optimization (PSO), viene determinata per ciacuna operazione il posizionamnto ottimale della base in termini di manipolabilità media durante l'esecuzione. Inoltre, tramite l’integrazione dell'algoritmo euristico Best Fit viene ottimizzato il posizionamento degli oggetti nelle scatole, minimizzando lo spazio inutilizzato e tenendo conto delle diverse categorie. L'applicazione dell'algoritmo ad un caso d'uso in ambito industriale, facendo uso dell'ambiente di simulazione di Tecnomatix Process Simulate, ha dimostrato come la configurazione con 20 particelle e 20 iterazioni del PSO possa portare ad un miglioramento del 12,18% nella manipolabilità rispetto ad una strategia a "base fissa", con un incremento del 6,37% nel tempo ciclo, giustificato dai vantaggi in termini di affidabilità. L’approccio pertanto si conferma efficace per contesti industriali reali, ottimizzando spazio, risorse e prestazioni robotiche.
Pre-deployment optimization of a robotic workstation with a movable linear axis for packing applications
Sonnino, Sole Ester
2023/2024
Abstract
In the last decades, industrial automation has seen an increasing employment of advanced robotic systems, such as mobile manipulators, to improve the efficiency, precision and safety of production processes. However, one of the most significant challenges for an effective usage of mobile manipulators lies in the optimal placement of their base, which is crucial in order to ensure high performance in terms of manipulability and execution time. This thesis proposes a pre-deployment optimization strategy for a robotic workstation equipped with a six-degrees-of-freedom manipulator mounted on a linear guide, used for packing operations of objects belonging to different categories. The output of the algorithm will serve as input for the real system, optimizing its performance from the start. The goal is to define an optimal sequence of pick-and-place operations, balancing manipulability and cycle time. Using the Particle Swarm Optimization (PSO) algorithm, the optimal base position associated to each operation is determined based on the mean manipulability measure during the execution. Additionally, by integrating the heuristic Best Fit algorithm, the placement of objects in boxes is optimized with the aim of minimizing unused space, while accounting for different categories. The application of the algorithm to an industrial use case, performed making use of the Tecnomatix Process Simulate environment, demonstrated that the PSO configuration with 20 particles and 20 iterations led to a 12.18% improvement in terms of manipulability compared to a "fixed base" strategy, with a 6.37% increase in cycle time, justified by the gain in reliability. This approach thus proves to be effective for real industrial contexts, optimizing space, resources and robotic performance.File | Dimensione | Formato | |
---|---|---|---|
2025_04_Sonnino_Tesi.pdf
accessibile in internet per tutti a partire dal 12/03/2026
Descrizione: Testo Tesi
Dimensione
3.82 MB
Formato
Adobe PDF
|
3.82 MB | Adobe PDF | Visualizza/Apri |
2025_04_Sonnino_ExecutiveSummary.pdf
accessibile in internet per tutti a partire dal 12/03/2026
Descrizione: Testo Executive Summary
Dimensione
1.51 MB
Formato
Adobe PDF
|
1.51 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/235701