Kshitiz Bansal

Kshitiz Bansal

Radar Engineer

Blue River Technology


I am Kshitiz Bansal, a Radar Engineer at Blue River Technology. I am working on bringing autonomy to agriculture. I finished my PhD in CSE department at UC San Diego where I was a part of WCSNG research group, advised by Prof. Dinesh Bharadia. My research interests include ML algorithms and their applications in wireless systems and sensors. My current work involves the fusion of radar sensing and computer vision with primary focus on autonomous driving systems. In my free time, I enjoy following football (the real football!) and reading books.

  • Radar Imaging
  • Computer Vision
  • Deep Learning
  • PhD in Computer Science Engineering, 2018 - 2023

    Univeristy of California, San Diego

  • BTech in Electrical Engineering, 2013 - 2017

    Indian Institute of Technology, Bombay


Radar Engineer Intern
Blue River Technology
Jun 2022 – Dec 2022 Sunnyvale, California

Responsibilities include:

  • Deployed an end-to-end radar based solution to detect objects in dusty conditions on farms for a self-driving tractor.
  • Developed a clustering and a machine learning based solution to identify useful objects in presence of the heavy ground clutter in radar point clouds.
Research Intern, AVANTE, Automotive Radar Team
Qualcomm Technologies
Jun 2020 – Sep 2020 California
  • Developed a deep learning based 3D object detection pipeline for sparse radar point cloud data.
  • Studied the trade-off between resolution and processing speed of radar’s bird-eye-view maps and the effect of using a point-based encoder.
Engineer, Advanced Connectivity and IOT team
Samsung Research Institute
Jun 2017 – Jul 2018 Delhi
  • Developed a deep learning model to perform predictive analysis on the data generated from network and system logs using network parameters like jitter and bandwidth for network fault prediction, security management and traffic optimization.
  • Obtained the Samsung Professional SWC Certification within 3 months of employment which is currently held only by less than 5% of around 1800 company employees.

Recent Publications

Quickly discover relevant content by filtering publications.
(2023). SHENRON - Scalable, High Fidelity and EfficieNt Radar SimulatiON. In RAL ‘23.

PDF Cite DOI Webpage

(2023). mmSpoof: Resilient Spoofing of Automotive Millimeter-wave Radars using Reflect Array. In S&P ‘23.

PDF Cite Webpage

(2022). VISTA: VIrtual STereo based Augmentation for Depth Estimation in Automated Driving. In ML4AD.

PDF Cite

(2022). R-fiducial: Reliable and Scalable Radar Fiducials for Smart mmwave Sensing. In arXiv.

PDF Cite

(2022). RadSegNet: A Reliable Approach to Radar Camera Fusion. In arXiv.

PDF Cite

(2020). Pointillism: accurate 3D bounding box estimation with multi-radars. In SenSys ‘20.

PDF Cite Code Dataset Video DOI Webpage