Jim Aldon D'Souza

• Achieved "key company milestone" with Algolux's first hands-free driving of a passenger vehicle on a test track with controlled traffic • Lead the planning and development of s/w modules for Algolux's autonomous driving software platform. ie algorithms for multi-sensor fusion, localization, mapping, prediction, planning, and control • Built the company's robotics roadmap to accompany the business strategy based on company's OKRs • Hiring manager for robotics engineer and motion planning engineer roles.

• multi-object tracking, 3D object detection from monocular video, monocular visual odometry, and lidar/camera sensor fusion • core components of state estimation pipeline, including visual-inertial odometry, visual odometry, and SLAM for autonomous driving • reverse engineering Tesla Autopilot in 2017 Model S P100d, data collection and collation pipeline, post-processing and comparison with internal ML models

Autonomous driving research. Focus on the vehicle's perception. This includes fusing data from various sources like LiDAR, Radar, camera sets, wheel encoders, IMUs and GPS to facilitate localization and obstacle mapping. Methods include conventional (non-linear estimation filters) as well as deep learning techniques.

Part of System Software team which fell under the Hardware Engineering arm of Nikon Metrology worldwide. Responsibilities include: • Designing a pipeline for real time scan data from a scanner to the application software • Writing efficient, time critical data processing components (including reflection and 3D speckle filters) along the scan pipeline • Code refactoring and QA

Reseach and develoment of localization and SLAM components for simulations as well as testing on real robot systems. My contribution addresses the localisation issues a robotic system faces when navigating and mapping a structurally symmetrical, and relatively featureless environment such as the ones encountered on a planetary surface.

Helping build a portable embedded platform for Visual Inertial Navigation using Extended Kalman Filter (EKF) and Simultaneous Localization and Mapping (SLAM) techniques. The project uses GPS-less sensor data from a monocular camera, accelerometer and an odometer. My work responsibilities include