Made easy with PIXKIT: hands-on From Zero to One, and One to Plenty in AV R&D
Autonomous driving is a complex endeavour, with difficult technical and regulatory challenges that require multidisciplinary teams to tackle. Robotics and AI represent the core of its development essentials; having systems interact in an intelligent way with their environment is one of the most important aspects.
To AV starters, it’s like finding a robotic body and matching it with eyes and brains.
10 days seem too little for such accomplishments — or not?
Why it is hard to start Autonomous Driving?
It’s mission possible for tech giants to get creative with their programming and engineering expertise, making it a race against the clock on promising projects. However, autonomous require an array of resources- from driving hardware like suitable vehicles with sensors to software like Apollo, Autoware and ROS- but finding the right ones can be challenging!
For instance, to get started, there is a range of “bodies” / ”body parts” for you to choose from:
1. Vehicles. The most popular ones include Tesla, Audi, Mercedes-Benz, Volvo, Honda, BMW, Ford, Nissan and Toyota which are equipped with sensors and AI algorithms capabilities.
2. Sensors. Lidar (Light Detection and Ranging), cameras, radars (radio detection and ranging), GPS, IMUs(Inertial Measurement Units)
3. Processors & controllers: microcontrollers, CPUs and GPUs
Then there is another range of “eyes and brains” to choose from, with the below most popular ones in the market:
1. Apollo. A comprehensive autonomous driving platform includes software for perception, planning, control, and simulation, as well as tools for data processing and analysis.
2. Autoware. An open-source autonomous driving software stack that is specifically designed for urban environments includes modules for localization, perception, motion planning, control ect.
3. ROS. Another open-source framework for robot software development for AV includes tools for building and managing complex robot systems, as well as libraries for perception, navigation, and control.
4. Waymo. It is designed by Alphbet Inc. and includes advanced sensors, software, and hardware for perception, mapping, localization, and control.
5. Tensorflow. An open-source machine learning framework that can be used to create and train neural networks for perception and decision-making in autonomous driving systems.
Wevolver has cast out a more detailed report in 2020, read the Autonomous Vehicle Report 2020 for more.
Finding the perfect robot for your project can be a quest — it’s like finding all of the body parts and matching them up with eyes, and brains…and then hopefully you have something that works! Although many options exist, adapting both software and hardware also requires sufficient time.
How to avoid going through such an arduous process?
PIX Moving has made life easier by providing R&D groups with complete AV robots along with instructions, which is called PIXKIT.
PIXKIT is a brand-new autonomous driving experience — a comprehensive R&D toolkit that allows you to build, test and iterate on your own custom AV solutions.
It comes with an integrative turnkey solution featuring sensors, computing hardware and software for both remote control and CAN agreement-based autonomy. Plus, Autoware is pre-installed so you’re always up-to-date! And don’t forget the lightweight skateboard chassis ensures safer performance while remaining light enough to move around easily. For faster learning times collaborate with global developers in this exciting new world of cutting-edge technology!
Who is PIXKIT designed for?
PIXKIT is mainly designed for engineers, teachers and students from educational institutions, and of course the early adopters from tech enterprises.
It features multi-level self-driving technology practices, real-world autonomous driving scenario simulations, and real-time data generation and feedback collection. PIXKIT helps AV research sector users master types of autonomous driving technologies, and improve their technical strength and innovation ability.
Core Features:
Integration of computing platforms, wire chassis, sensors and autonomous driving systems for users to focus on algorithm development;
2. Lower development thresholds: provides maps and sensors to calibrate toolchains;
3. Open source: all source code and provide detailed documentation to make development more responsive;
4. Up-to-Date technologies: keep syncing with the most forefront algorithms and technologies
5. Drive development thanks to ROS/Atoware community resources
6. Provide different ODD examples to extend Autoware to different scenarios (subsequent support)
User Case
University of UPS (Universidad Politécnica Salesiana) in Ecuador
Recently, the University of UPS (Universidad Politécnica Salesiana) in Ecuador announced the launch of the “Autonomous Vehicles # ANTA” project, implemented by the Autonomous Driving Engineers and Transport Engineering Research Group.
To celebrate the founding of the project, UPS University conducted a launch ceremony to which more than 100 Ecuadorian leaders in research and business were invited.
In this Smart Mobility project, the PIXKIT provides students with a high-quality and efficient hands-on experience, accelerating their understanding and mastery of technological innovations of autonomous driving technologies, and injecting new impetus into the development of smart transportation in Ecuador.
2. SpringCloud in South Korea
About PIX
With the goal of rebuilding the city with autonomous mobility, PIX Moving aspires to innovate not only new-generation cars, but a chain of tools for automobile design and production. PIX Moving Space (mSpace) will be the main fully-autonomous driving product representing the third-generation automobile revolution. It will provide on-demand service in urban areas and usher in a new era of “Mobility-as-a-Service” (MaaS).