Frequently asked questions
Jump to a question:
What is the QCar?
The Qcar is a 1/10 scale model fully open-architecture autonomous car. As a teaching device it is designed to introduce students to a wide range of topics in guidance, navigation and control of self-driving cars accompanied by teaching curriculum. As a research platform it is designed to allow year-round and rapid validation of control, autonomy, artificial intelligence and machine learning algorithms developed by researchers before deployment on full scale models. The Qcar deploys a powerful Nvidia TX2 quad-core processor onboard and is equipped with Lidar, front facing RGBD camera, sonar, 4 additional cameras to provide 360-degree view, onboard IMU, wheel encoders and tachometers as well as wifi for communications. This works aims at developing a novel stereo vision sensor based on deep neural networks, that can be used to produce a feedback signal for visual based localization and mapping iwith the Qcar. Two deep convolutional neural networks attached to the stereo camera in the drone are trained to perform stereo triangulation to calculate the distance and orientation of objects of interest.
What is YOLO3
YOLO3 is one of the three types of object detectors used in Nueral Networks.
YOLO works completely different than most other object detection architectures. Most methods the model to an image at multiple locations and scales. High scoring regions of the image are considered detections. Yolo, on the other hand, applies a single neural network to the full image. The network divides the image into regions and predicts bounding boxes and probabilities for each region. These bounding boxes are weighted by the predicted probabilities. Through this technology we can detect over 6000 cateewgories of items including if someone is wearing a mask or not
