• Working on microservices that provides state estimation and state prediction for robots.

• Implementing production level code.

• Working on RFID readers.

• Using softwares and technologies like Docker, Kubernetes, C#, Message brokers, Redis.

• Working on addition/modification of data APIs over UDP for IMU and UWB range sensor ROS driver and firmware.

• Analyzing multiple IMUs using statistical methods and algorithms to drive choice of IMU for the product.

• Performed localization performance evaluation for new in-house localization hardware and automated the navigation module.

• Worked on design and implementation of algorithm for obtaining good UWB range sensor data.

• Integrated third party ROS packages and libraries to the existing code base.

• Modified docker compose files, built custom docker images and loaded the images on hardware to test the whole pipeline.

• Modified existing Shell script on Linux machines to add functionality for using Linux services, running ROS launch files and checking status of the processes.

• Planned, created, executed test cases and provided reports for software defects through JIRA in an agile software development cycle. Released and maintained Confluence pages following Software team procedures.

• Used Debugging and profiling tools (gdb, valgrind)

Designed and developed a tool for field deployment engineers for evaluation of Odometry, IMU and Range sensor data used for Localization using ROS framework. Improved understanding of Gazebo simulation (URDF, sensor plugins, controller plugins), sensor fusion and different type of robot models (Differential, Ackermann).

• Worked on wearable biosensor, extracted information from different sensors and used sensor fusion techniques to combine them to get Heart Rate, Respiration Rate, Activity Type, Activity Level, Posture and other things.

• Implemented MATLAB scripts to deal with time series data and find out false positives and positive errors.

• Used Fluke simulators to mimic humans and collected data from it. Implemented Matlab scripts to generate synthetic data from physiological package and verified the core algorithms used to measure heart rate and respiration rate and wrote verification protocols for all the test cases.

• Wrote a data reader module using C++ and Windows environment, whose function was to basically convert the raw files (binary data) to a human readable data (ASCII), to debug the data packet, check the consistency of the waveform and other things.

• Used agile methodology, which proves to be advantageous because of reduced risks, faster ROI, higher customer satisfaction and many more reasons.