Goal: To Advance Career | Salary: ₹50 lakh | Nanodegree | Proficiency in Python, C++, Linear Algebra and Calculus | Taught by Udacity in collab. with Waymo and Mercedes Benz | Duration: 5 Months
This program helps you specialise in high-value in-demand driverless car technologies. You learn the skills and techniques used by self-driving car teams at the most advanced technology companies in the world, and prepare for jobs such as System Software Engineer, Deep Learning Engineer, Vehicle Software Engineer, Localization and Mapping Engineer, etc.
Please note, you should have good proficiency in Python, C++, Linear Algebra, Calculus and Statistics before you take up this program.
As such, we suggest that you should take up the Intro to Self-Driving Cars Nanodegree program before this program so you learn the fundamentals of driverless car technologies before you take up this specialisation.
What You’ll Learn
- How camera sensors work and how to process raw digital images before feeding them into neural networks
- How to build convolutional neural networks using TensorFlow and learn how to classify and detect objects in images
- How Sensor Fusion works, how to detect objects such as vehicles in a 3D LiDAR point cloud using a deep-learning approach, and how to fuse camera and LiDAR detections and track objects over time
- About robotic localization, from one-dimensional motion models up to using three-dimensional point cloud maps
- About path planning routes and how to react when emergencies arise. This section is taught by the Mercedes-Benz Vehicle Intelligence team
What You’ll Earn
- ₹50 lakh per year is the average starting salary for a Driverless Car Engineer in India (Source: businesstoday.in, businessinsider.in)
- $200,000+ per year is the average salary for a Driverless Car Engineer in the US (Source: businesstoday.in, businessinsider.in)
Details
This program will teach you everything you need to become a driverless car engineer.
Here are the skills you will learn:
- Computer Vision: Point Cloud Data, Non-Linear Motion Tracking, Point Clouds, Object Tracking, Object Clustering, Digital Image Keypoint Descriptors
- Robotics: Sensor Fusion, Radar Clutter Thresholding, Object Motion Models, Autonomous Vehicle Fluency
- Data Structures and Algorithms: Kalman Filters, RANSAC Algorithm, Extended Kalman Filters
- Autonomous Vehicles and Robots: Radar, LiDAR
- Trigonometry: Basic Trigonometry
- C++: Point Cloud Library
Applied Learning Project
The program offers you intensive hands-on projects on real-world driverless car problems. These projects equip you with invaluable skills across a wide array of critical topics, including computer vision, sensor fusion, localisation, motion control, and more.
Over the course of the program, you will, among other things:
- Create a convolutional neural network to detect and classify objects using data from the Waymo Open Dataset
- Load and preprocess 3D LiDAR point clouds, and use deep learning to detect and classify objects. Evaluate and visualize objects, including calculating key performance metrics
- Solve a challenging multi-target tracking task by fusing camera and LiDAR detections
- Use either of the two scan matching algorithms, ICP or NDT, to align point cloud scans from the CARLA simulator and recover the position of a simulated car with LiDAR
- Implement two of the main components of a traditional hierarchical planner: the behaviour planner and the motion planner
- Design a PID controller to perform vehicle trajectory tracking so you understand the power and the limitations of the PID controller and how feedback control is used in practice
