Spearheaded the development of stereo vision based 3D object detection for Grimme autonomous harvesters.  

Conducted model benchmarking on custom datasets and hardware performance profiling of SOTA object detection models for agricultural scenarios. 

Used model quantization methods to achieve a 5x increase in object detection inference speed using ONNX and HailoRT for the Hailo8 processor on a x86 host edge computer. 

Guided the development of a custom dataset for image segmentation tasks required for field row navigation tasks for autonomous harvesting.  

Developed a perception pipeline for safe mobile robot operation in construction environments with a focus on stereo vision based 3D object detection and LIDAR point cloud compression.   

Designed and deployed a closed-loop LIDAR tilt system to increase the vertical field of view by 100% for close range applications. The ROS package was developed using C++ and deployed on an NVIDIA Jetson Xavier Platform. 

Conducted literature research and documented the existing object detection models as well as point cloud compression methods.  

Utilized machine learning and sensor fusion techniques to develop a calibration algorithm for the dynamic payload estimation system of a self loading cement mixer truck for the mining and construction industry. 

Implementation of calibration and weight estimation algorithms using C++ on edge devices (ARM Architecture).

Sensor data acquisition and analysis using python. 

Use case testing and software documentation. 

Achieved weight estimation accuracy of 98.5% with a maximum payload of 800kg in dynamic conditions.

Designed and developed control software (C++) for precise liquid dispensing for smart composite manufacturing at the Department of Aerospace Engineering 

Designed a sensor module to automatically detect composite fabric undulations. 

Designed and Developed a testbed system to dispense Epoxy-Hardener Mixture with dispersed Multiwalled Carbon Nanotubes.

Achieved an accuracy of up to 10 microliters using a MEMS-based flow sensor and precise pressure control. 

Worked on the development of an AS/RS Mobile Robotic System for medium and large-scale warehouses. 

Worked on the mechanical design of an autonomous warehouse rover for pick and place applications of crates up to 40Kgs in weight.

Designed and developed a lifting mechanism for autonomous rovers to lift up to 1 tonne of payload weight.

Partnered up with various domestic and international vendors for part procurement. 

Founded and managed a team of 6 from various engineering disciplines to design and develop the Modulus Series of Gantry Robotic  Systems for the MSME sector and Prototyping Labs. 

Presented the product business plan at IIM Bangalore and led the team to the top 30 teams in India at the Texas Instruments India Innovation Challenge. 

Raised a total of $ 13000 for prototype development 

Deployed Modulus 1 and Modulus R for consumer testing and partnered up with local industry for product validation.

Worked on the enhancement of the engineering and product design to implement a modular product architecture using methods such as Design Structure Matrices at the Social Sustainability Lab under the guidance of Prof Pankaj Upadhyay. 

Worked with industrial designers on the product concept development of Modulus X. 

Presented and published our research and case study of modulus 1 at the ICORD 2019 conference held at the Indian Institute of Science. 

Worked on the Development of an Ontology and knowledge base for aircraft assembly using protege under the guidance of Prof B. Gurumoorthy. 

Learned various aspects of product design and design thinking approaches to creating a robust product for the industry.