Developing motion planning algorithms of autonomous delivery robots for smart cities

Nurasheva, Jamilya (2025) Developing motion planning algorithms of autonomous delivery robots for smart cities. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

The widespread development and integration of artificial intelligence and robotics technologies in all areas of our lives, including logistics, has contributed to the emergence and development of autonomous delivery robots (ADRs). Courier robots can reduce delivery costs and carbon emissions and improve the efficiency of the ‘last mile’. However, developing ADRs is a complex task, as many details must be taken into account, such as static and dynamic obstacles on the terrain, the unpredictable behaviour of pedestrians, cyclists and other road users, as well as the complex infrastructure of the city.

This work is dedicated to the study and comparative analysis of three approaches to motion planning: the classic A* algorithm, the Social-LSTM machine learning model, and the A*+Social-LSTM hybrid. Two open, publicly available datasets were used to train and test the approaches: nuScenes and Stanford Drone Dataset (SDD).

The ADE, FDE, and RMSE metrics were selected to evaluate the performance of the algorithms. The results of the study showed that the classic A* method performs well in static conditions, while the Social-LSTM-based model demonstrated high accuracy in predicting the agent's trajectory but did not take into account the physical infrastructure of the map. The hybrid model performed best, showing a balance between adaptability and structure.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Vamadevan, Arundev UNSPECIFIED
Uncontrolled Keywords:	Motion planning; autonomous delivery robots; Social-LSTM; A*,; hybrid methods; trajectory prediction
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	02 Jun 2026 11:42
Last Modified:	02 Jun 2026 11:57
URI:	https://norma.ncirl.ie/id/eprint/9336

Actions (login required)

View Item