In my role as a Software Engineer at Robotic Systems, I was part of the development team responsible for enhancing our flagship autonomous delivery robot. This project aimed to improve the robot's ability to navigate complex urban environments while carrying parcels to customers' doorsteps.
The fundamental technologies that powered our robot included a Linux-based control system and several open-source tools. We primarily used Robot Operating System (ROS) for creating complex behaviors and OpenCV for computer vision tasks, like object detection and path planning.
I was responsible for improving the robot's obstacle avoidance abilities. Leveraging sensor fusion, combining the data from the LiDAR sensors and stereo cameras, I developed algorithms for creating a dynamic map of the robot's surroundings. Using this map, the robot could predict and avoid obstacles, thereby increasing its effectiveness.
The project was a success, with our upgraded robot demonstrating a 20% reduction in delivery times and a marked increase in safety metrics. These improvements positively impacted our customers' satisfaction, therefore contributing to our bottom line.