Simulation for GEMstack (Autonomous Driving Stack for GEM Vehicle)

Sai Rohit Muralikrishnan, Harish Kumar Balaji, Vikram Raj, Praveen Kalva, Daniel Zhuang

Github

Developed core simulation infrastructure for the GEMstack autonomous vehicle platform as part of the CS588 Simulation Group. Developed a YAML-based Gazebo simulation framework for GEMstack, enabling dynamic scenario creation and realistic 3D environments to test the full perception, planning, and control stack.

YOLOPose

Sai Rohit Muralikrishnan, Harish Kumar Balaji, Vikram Raj, Yudai Yamada

Github

A take on finding 6 Degrees of Freedom which a robot arm can use to pick up an industrial warehouse part using computer vision.

Music Generation: Emotion-Adaptive

Sai Rohit Muralikrishnan, Aditya Pillai, Dhruv Agarwal

Github

This project implements an Emotion-Aware Music Generator (EAMG) that turns natural-language prompts into emotion-aligned music in real time. It uses a LoRA-tuned DistilBERT to classify emotions, maps them to musical parameters using music theory, and generates MIDI via a transformer model with KV caching for low-latency output.

Autonomous Drone Racing with Vision-Aided Minimum Snap Trajectory Generation and MPC-Based Velocity Control

Sai Rohit Muralikrishnan, Murali Karteek, Abhiram Annaluru, Ayush Tarachandani

Github

Developed a vision-guided system in AirSim for drone racing, featuring minimum snap trajectory generation with spline interpolation. Implemented MPC for planar control and PID for elevation to ensure precise gate navigation and real-time adjustments.

FROM VISION TO UNDERSTANDING : Leveraging VLMs to enable autonomous driving decisions

Sai Rohit Muralikrishnan, Murali Karteek, Brijesh Muthumanickam, Sridhar T

Github

Integrated advanced VLMs like SpatialBot with multi-camera setups to enhance autonomous driving decisions. Developed pipelines for object tracking (YOLOv8, SAM2) and ego-vehicle motion estimation (optical flow, depth mapping), optimizing real-time navigation in dynamic environments.

SketchUR3

Sai Rohit Muralikrishnan, Mei Han

Github

Implemented a ROS-based framework integrating OpenCV for contour detection, coordinate transformations for workspace mapping, and motion planning (MoveJ and MoveL) to replicate images with high precision.

Smart Caravan

Sai Rohit Muralikrishnan, Rathan Nidhi Kumaran, Shalen Brisil

Ideated an autonomous caravan solution incorporating LIDAR, RADAR, GPS, and biometric security to enhance safety and automation. Designed a dual-energy system and space-saving amenities for off-grid travel, earning finalist recognition among 30,000+ participants in the Techgium 2021 competition.

IoT-Based Water Quality Monitoring and Prediction

Sai Rohit Muralikrishnan

PDF

Developed an IoT system using pH, temperature, and conductivity sensors with neural networks to monitor and predict water quality in real time, achieving 88% accuracy and enabling proactive measures for household and environmental applications through cloud integration.