AI is enabling robots to do more. As robots are used indoors to give people directions and deliver things, it is necessarily to enable them with the proper tech to navigate indoor environments safely. Four high school NVIDIA interns, Team CCCC (ForeSee) have used deep learning to add hazard avoidance to robots without spending lots of fortune.
As explained on NVIDIA’s website:
To detect complex obstacles using a single camera, the team trained the DeepLab V3 neural network architecture with ResNet-50 v2 as a feature extractor to segment free space (space not occupied by obstacles) on the ground in front of the robot. This enables the robot to travel through dangerous environments, which couldn’t be done previously with just a 2D LiDAR.
The team used TensorBoard to monitor network training on NVIDIA Tesla V100 GPUs. The final robot consists of a 2D LiDAR, webcam, and an NVIDIA Jetson TX2 with ROS.