Agent-based COmputational Modeling Lab

Name of the Lab: Agent-based Computational Modeling Lab

 Principal Investigator:

Syed Waqar ul Qounain
Punjab University College of Information Technology (PUCIT),
University of the Punjab, Old Campus, the Mall, Lahore.

Phone: +92-42-111923923-ext-547, Email: swjaffry@pucit.edu.pk

CO-Principal Investigators:

Muhammad Murtaza Yousaf

Punjab University College of Information Technology (PUCIT),
University of the Punjab, Old Campus, the Mall, Lahore.

Phone: +92-42-111923923-ext-541, Email: murtaza@pucit.edu.pk

Shahzad Sarwar

Punjab University College of Information Technology (PUCIT),
University of the Punjab, Old Campus, the Mall, Lahore.

Phone: +92-42-111923923-ext-508, Email: s.sarwar@pucit.edu.pk

Mian Muhammad Mubasher

Punjab University College of Information Technology (PUCIT),
University of the Punjab, Old Campus, the Mall, Lahore.

Phone: +92-42-111923923-ext-528, Email: mubasher@pucit.edu.pk

Goal: design and utilize agent based modeling and simulation (ABMS) techniques for better understanding of the world around and within us, and to craft state-of-the-art computational artifacts based on research findings, for the wellbeing of humanity.

Scientific Objectives:

  • Design computational artifact to seamlessly log driver’s observable driving behavior
  • Identify, investigate relationship between driver’s measurable characteristic (physiological, psychological etc.) and driving behavior
  • Formulate relationship between driver’s measurable characteristic and driving behavior using mathematical and computational formalisms as the humanistic driver agent (HDA)
  • Analyze, and validate HDA using human-based experiments
  • Predicting future behaviors and errors of drivers, in different driving conditions using the HDA
  • Design driver training program tailored for the improvement of driver’s behavior Based on driver specific error, highlighted by the HDA
  • Design in-vehicle driver support system using HDA

Description of Lab:

Recently, computational modeling and simulations has emerged as a third way of doing scientific study besides induction and deduction. Computational modeling is to represent and augment our understanding of the world around us. With the passage of time, the world is becoming more and more complex place, to live. To deal with this complexity, we are building systems that are in fact leading to a further increase in the complexity of the world. To understand these complex systems in their totality is no more a task of human mind. We need computational techniques to represent these complex systems as models and to use computational power to see their outputs under different scenarios. The answers to the following questions would be non-trivial for example to see a) how road crossing behavior of different pedestrians will generate traffic congestion on non-signalized interactions, b) how the traffic congestion will affect driver’s behavior, c) how does the change in behavior of driver will affect road accidents, and d) what would be the effect of road accidents on national economy. Similarly, to address complex questions, we need computational modeling techniques and the power of computer to help us. Hence, the focus of the proposed Agent-based Computational Modeling (ACM) laboratory is to

design and utilize agent based modeling and simulation (ABMS) techniques for better understanding of the world around and within us, and to craft state-of-the-art computational artifacts based on research findings, for the wellbeing of humanity.

According to world health organization (WHO), worldwide 1.25 million people die in road accidents every year. If the trend continues then traffic accidents will be 7th leading cause of death by the year 2030 in the world. In last twelve years, 1.28 million accidents took place in province of the Punjab, reported by Rescue 1122 emergency service. According to several studies, road accidents not only cause loss of human life but also cause huge financial loss amounting up to five percent of gross domestic product (GDP) in a developing country. As a developing country, Pakistan needs an immediate action plan to cut huge financial bearing and to meet global and local targets which are set by the UNO SDG 2030. This challenge would be dealt by addressing at multiple levels through cross disciplinary research and development (R&D). In this regard, the ACM Lab has envisioned a holistic prospective to address this problem. To proceed further, the fundamental requirement is to understand, completely or partially, behavior of human drivers. Hence, a humanistic driver agent (HDA) would be designed and validated by involving human drivers in experiments. The design of HDA would be inspired by findings of studies in various domains such as driver psycho-physiology, traffic psychology, and computational modeling.  In other three projects, the HDA would be used to design computational artifacts which includes in-lab infrastructure for training and assessment of drivers, real time in-vehicle driver support device that would be customizable according to behavior of the driver, and a simulation platform to evaluate and compare different design of transportation policy, and infrastructure interventions. These computational artifacts would help to reduce overall life loses on road and the cost of transportation management and planning. During the first three years, the ACM Lab would primarily focus on intelligent transportation systems (ITS).

Applications Domains:

    • In-Lab driver training environment
    • In-Vehicle drivers support
    • Anticipation of human drivers maneuvering in driverless vehicles
    • Transportation engineering and management
    • Computer games