Agent-based modelling and simulation of sidewalk delivery robots’ interaction with pedestrians

With the rise of e-commerce, consumer demands for parcel delivery have changed. COVID-19 further fueled this demand due to restrictions and people’s preferences for fewer interactions. With the increase in consumers’ delivery demands, the cost and complexity of meeting those parcel-delivery demands have been increased as well. There have been many disruptions to solve this problem, but sidewalk-delivery robots stand out because of their efficiency and ability to navigate through sidewalks. The deployment of sidewalk robots seems to be more appropriate than others due to multiple reasons like easy operation mode, cost, and safety concerns. The sidewalk delivery robots share resources that were primarily built for human-only like sidewalks. To study and analyze human-robot interaction, their movement behavior, and the benefits they can offer, we will require to build a digital twin of a large spatial area. A high-fidelity map will ensure that we test pedestrians and delivery robots close to the conditions they will be operating in the real world. The proposed simulation suite will offer tools and methods to test these interactions with different conditions. Simulation results will confirm that such interactions and interventions can improve the safety and mobility of sidewalk users.

Faculty Supervisor:

Gunho Sohn

Student:

Partner:

Esri Canada Ltd

Discipline:

Computer science

Sector:

Information and cultural industries; Professional, scientific and technical services

University:

York University

Program: