About Me

I am a fourth year Computer Science PhD student at the University of Southern California. I am part of the CPS-VIDA group, and my advisor is Prof. Jyotirmoy Deshmukh. My research interests lie in the design and verification of controllers for cyber-physical systems.

My research involves:

• Incorporating formal methods in the design of controllers for autonomous systems. This includes:
  • Design of reward functions for reinforcement learning agents, given temporal logic task specifications.
  • Controller synthesis for time-sensitive and safety-critical tasks.
• Fault detection and verification of perception-based control systems, especially in the context of autonomous vehicles.
  • Use of runtime monitors to detect malfunction in perception systems, especially in the context of multi-object detection and tracking.

Before this, I received my B.S. in Computer Engineering from the University at Buffalo in 2018, where I also worked with Prof. Karthik Dantu at the Distributed Robotics and Networked Embedded Systems Lab.

Selected Publications

• Model-Free Reinforcement Learning for Symbolic Automata-encoded Objectives

  Anand Balakrishnan, Stefan Jaksic, Edgar A. Aguilar, Dejan Nickovic, and Jyotirmoy Deshmukh.

  In arXiv/2202.02404, February 2022.

• PerceMon: Online Monitoring for Perception Systems

  Anand Balakrishnan, Jyotirmoy Deshmukh, Bardh Hoxha, Tomoya Yamaguchi, and Georgios Fainekos.

  In Runtime Verification, October 2021.

  arxiv

• Model-Based Reinforcement Learning from Signal Temporal Logic Specifications

  Parv Kapoor, Anand Balakrishnan, and Jyotirmoy V. Deshmukh.

  In arXiv/2011.04950, November 2020.

• Augmenting Visual SLAM with Wi-Fi Sensing for Indoor Applications

  Zakieh S. Hashemifar, Charuvahan Adhivarahan, Anand Balakrishnan, and Karthik Dantu.

  In Autonomous Robots, December 2019.

• Structured Reward Shaping Using Signal Temporal Logic Specifications

  Anand Balakrishnan, and Jyotirmoy V. Deshmukh.

  In 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), November 2019.

  PDF

• Using Logical Specifications of Objectives in Multi-Objective Reinforcement Learning

  Kolby Nottingham, Anand Balakrishnan, Jyotirmoy Deshmukh, Connor Christopherson, and David Wingate.

  In arXiv/1910.01723, October 2019.

• Specifying and Evaluating Quality Metrics for Vision-based Perception Systems

  Anand Balakrishnan, Aniruddh G. Puranic, Xin Qin, Adel Dokhanchi, Jyotirmoy V. Deshmukh, Heni Ben Amor, and Georgios Fainekos.

  In 2019 Design, Automation Test in Europe Conference Exhibition (DATE), March 2019.

  PDF

For most up-to-date publication information, check out my DBLP or Google Scholar pages.