SMART: Future Urban Mobility (FM)

client logos

Assortment of our research

Featured: Research Paper

Title: Demand Calibration of Multimodal Microscopic Traffic Simulation using Weighted Discrete SPSA

Authors: Oh, S., Seshadri R., Azevedo C. L., & Ben-Akiva M. E. (2019)

Abstract: This paper presents a stochastic approximation framework to solve a generalized problem of off-line calibration of demand for a multimodal microscopic (or mesoscopic) network simulation using aggregated sensor data. A key feature of this problem is that demand, although typically treated as a continuous variable is in fact discrete, particularly in the context of agent-based simulation. To address this, we first use a discrete version of the weighted simultaneous perturbation stochastic approximation (W-DSPSA) algorithm for minimizing a generalized least squares (GLS) objective (that measures the distance between simulated and observed measurements), defined over discrete sets.

Read more
perspective phone

Intra-FM Training




Malika is a postdoctoral scholar with Autonomous Vehicle group at Singapore-MIT Alliance for Research and Technology (SMART). She was awarded SMART Scholarship in 2017 for her research proposal on autonomous fleet management using heterogeneous robots and path planning for self-driving cars. She received a PhD degree in Computer Science from McGill University, Canada in early 2017. In 2016, her work on "Multi-Target Rendezvous Search", was nominated as the finalist for the best paper award at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). A start-up proposal based on her work, titled, “Multi-Agent Rendezvous on street networks”, won her the NSERC Canadian Field Robotics Network, Strategic Network Enhancement Initiative Award in 2015.

Robot Operating System - A hands on experience

Instructor: Dr. Malika Meghjani

In this tutorial, I provided an overview of the Robot Operating System (ROS), an open-source robotics middleware. ROS is a widely used software framework for integrating heterogeneous sensors and hardware through low-level device control, message passing between processes and package management. The fundamentals of ROS were discussed through understanding of debugging and command-line tools along with catkin build and launch systems. Each of the basic concepts of ROS, such as topics, nodes and messages, was learned through simultaneous programming with the participants. A hands on interfacing experience was also provided to the participants with a small scale autonomous vehicle.




Shashwat VERMA was an intern with Singapore-MIT Alliance for Research & Technology Centre. He is currently a Software Engineer working on the perception in autonomous vehicles under the Autonomous Mobility on Demand project.

Deep Learning

Instructor: Shashwat VERMA

In this course, the fundamental elements of deep learning, what it means, how it works, and develop code necessary to build various algorithms such as deep convolutional networks, variational autoencoders, generative adversarial networks, and recurrent neural networks. A major focus is on the application of these algorithms and explore other alternative use cases.


Our Latest Achievements


Events & Seminars

Lunch Seminar April 25
    Context and Intention Aware Planning for Autonomous Vehicles and...
    Abstract: In this talk, I will present a context and intention aware motion planner for autonomous driving in urban environments. Unlike highways, urban environments require...
Lunch Seminar June 12
    Transportation Network Flow Game: Proactive Planning...
    Abstract: The increasing number of automated devices associated with intersection management (e.g., traffic light controllers) in urban transportation...

1 CREATE Way, #09-01/02 CREATE Tower;
#01-13 Enterprise Wing, Singapore: 138602