Master Thesis

Robust ego-motion estimation of an Autonomous Delivery Device

Work default illustration

Information

  • Started: 01/02/2022
  • Finished: 14/07/2022

Description

The amount of goods to be delivered in metropolitan areas will increase dramatically in the next few years. Deliveries are more frequent and fragmented, especially because of a skyrocketing use of e-commerce. Today's logistic operations in city centreslead to very negative effects (e.g., increase in traffic congestion; safety problems for pedestrians, bikers and deliverers; or air and noise pollution) and to tackle these challenges, there are several ongoing efforts to design new fully autonomous electric delivery devices.

This thesis proposal aims at designing and implementing the required perception system to estimate the ego-motion of an autonomous delivery device. Particularly, the main solution will explore the fusion of the information provided by the robot's wheel odometry, geolocalization (GPS), LiDARs, cameras and IMU. Further, the designed architecture will provide diagnostics of the internal states of the sensors and algorithms to account for the system's health.

The selected candidate will join the research team of the mobile robotics line working on an ADD vehicle at the Institut de Robòtica i Informàtica Industrial, CSIC-UPC. The student will potentially be supported by a UPC-INIREC fellowship throughout the semester and a publication of a scientific paper will be targeted together with the thesis report.

Focused for students of the following degrees (but not limited to):
• Màster Universitari en Automàtica i Robòtica
• Doble Màster Universitari en Enginyeria Industrial i Automàtica i Robòtica

The work is under the scope of the following projects:

  • LOGISMILE: Last mile logistics for autonomous goods delivery (web)