Autonomous Vehicle Companies and Their ML
The race in the autonomous vehicle space has resulted in a large number of companies vying for a position at the top. Here's what some of the main ones are doing.
Created on September 6|Last edited on July 16
Comment
The development of autonomous vehicles is a highly competitive landscape. A large number of companies are working on the challenge, including many of the world's largest automakers. There is a lot of investment flowing into the area, and the race to develop the technology is heating up.
The main players in the autonomous vehicle space include Tesla, Uber, Waymo, and a number of traditional automakers such as BMW, Mercedes-Benz, and Ford. These companies are all vying for a piece of the pie, and the competition is fierce.
There are a number of different approaches to autonomous vehicle development, and each company has its own strengths and weaknesses. It is still early days, and it is hard to say who will come out on top, but whoever can develop the technology and get it to market first will have a major advantage.
Here's what we'll be covering:
Table Of Contents
Autonomous Vehicles ManufacturersTeslaRivianZooxGeneral MotorsWaymoCompanies Making ML Datasets for Autonomous VehiclesLyftBerkleyFordMicrosoftOxford RobotcarnuScenesToyota/Karlsruhe (KITTI-360)Companies making Machine Learning Models for autonomous vehiclesLevel 5comma.aiBaidu (Apollo)Companies Making Sensors for Autonomous VehiclesContinentalInnovizLuminar TechnologiesFinal Thoughts About The Future Of Autonomous VehiclesRecommended Reading
Autonomous Vehicles Manufacturers
There are many different companies that are trying to get a piece of the autonomous vehicle manufacturing pie. Some of the top companies include Tesla, Rivian, Zoox, GM, and Waymo. Each company has its own unique approach to AV manufacturing, and it will be interesting to see how the landscape changes in the coming years.
Tesla
Tesla is an American electric vehicle and clean energy company based in Palo Alto, California. Tesla's first product was the Tesla Roadster, an electric sports car that was launched in 2008. The company's second product, the Model S, is a full-fledged electric sedan that was introduced in 2012. The Model S was followed by the Model X, a full-sized SUV, in 2015. Tesla's latest vehicle is the Model 3, a mass-market electric sedan that was launched in 2017.
In addition to selling electric vehicles, Tesla is also a leader in autonomous vehicle technology. The company's Autopilot system is a suite of advanced driver assistance features that allow Tesla vehicles to drive themselves in certain conditions. Tesla is also working on fully autonomous vehicles that will not require any input from a human driver.
These vehicles are still in development and are not yet available to consumers.
Rivian
Rivian is an electric vehicle manufacturer, developing autonomous vehicle technology. The company was founded in 2009 by R. J. Scaringe, who is also the current CEO. Rivian is headquartered in Irvine, CA, and has a manufacturing facility in Normal, Illinois.
Rivian's first product is an all-electric sport utility vehicle, the R1S, most deliveries for which are scheduled between October and December 2020. The R1S will have a range of 316 miles (508 km) and a starting price of $72,500. The company is also developing an all-electric pickup truck, the R1T.
Rivian is also working on autonomous vehicle technology called "Driver +".
Zoox
Zoox is a company specializing in manufacturing cars and autonomous vehicle technology. The company was founded in 2014 by Tim Kentley-Klay and Jesse Levinson. Zoox's mission is "to make personal transportation safer, cleaner, and more enjoyable—for everyone."
The company was bought by Amazon in 2020 and is currently most famous for its work on robotaxis.
General Motors
General Motors is one of the world’s largest automakers. The company has been making cars and trucks for over a hundred years. GM is also a leading producer of electric vehicles. The company’s Chevy Bolt EV was the first mass-produced electric car to have a range of over 200 miles per charge.
GM is now working on self-driving cars. The company plans to launch a fleet of autonomous vehicles in 2019.
Waymo
Waymo is an autonomous vehicle technology company and a subsidiary of Google's parent company, Alphabet. Waymo was founded in 2009 as the Google Self-Driving Car Project. It was later renamed Waymo in 2016. Waymo is headquartered in Mountain View, CA.
The company has been testing self-driving cars on public roads since 2015. In 2017, Waymo launched a pilot program in Phoenix, Arizona, where select members of the public could hail a self-driving car using a smartphone app. The program has since been expanded to other cities.
Companies Making ML Datasets for Autonomous Vehicles
There are many factors to consider when making an autonomous vehicle, but one of the most important is the quality of the datasets used.
Autonomous vehicles rely on datasets to learn to make decisions about how to navigate, and if those datasets are of poor quality, the vehicles will likely make poor decisions as well. This could lead to accidents or other problems. Therefore, it is essential that the datasets used to train and test autonomous vehicles are of the highest quality possible.
Here's a list of companies and institutions working on providing these datasets for AVs.
Lyft
The company famous for its ridesharing service is also working at the forefront of the development of machine-learning-based approaches to autonomous driving. The company has most notably open-sourced its "Lyft Motion Prediction for Autonomous Vehicles" dataset on Kaggle and hosted a competition with $30,000 in prize money.
Berkley
The next entity on our list is not a company but in fact, is Berkley University with its "Berkley Deep Drive" (BDD) dataset. This is a large dataset with over 100K of videos collected from over 50K rides with different evaluation tasks within it. The tasks include lane detection, instance segmentation, panoptic segmentation, semantic segmentation, object detection.
Ford
The Ford Autonomous Vehicles Dataset is the result of a year-long experimental deployment of autonomous vehicles in urban environments (2017-2018).
The dataset captures the complexity of the environment and the variety of conditions the vehicles encountered. It can also be used to develop and test algorithms for autonomous vehicles, multi-agent systems, and intelligent transportation systems.
The dataset contains raw data from all sensors, calibration values, pose trajectory, ground truth pose, and 3D maps. All data is available in Rosbag format.
Microsoft
Next up on our list is Microsoft's AirSim dataset, providing realistic-looking simulations for obtaining synthetic data for such tasks as navigating drones in the air and cars on the ground. In 2017, the original AirSim was replaced with a new project targeting similar tasks.
Oxford Robotcar
The Oxford RobotCar Dataset is a large, long-term dataset of driving data captured in Oxford, UK. The dataset contains many different combinations of weather, traffic, and pedestrians, as well as the different layouts of the roads, including constructions and roadworks appearing. This makes it an excellent resource for developing and testing autonomous driving systems.
nuScenes
The special thing about this dataset is that it is the first one that was captured using the full set of AV sensors (6 cameras, 5 radars and 1 lidar, all with full 360-degree field of view). nuScenes has 1000 scenes, lasting under a minute with annotated 23 3D object detection classes and 8 other attributes. And it's a public dataset that can be used by anyone.
Toyota/Karlsruhe (KITTI-360)
The KITTI-360 dataset follows such tasks of interest as stereo, optical flow, 3D tracking, visual odometry, and 3D object detection.
The interesting thing about the dataset is that it provides the data from two high-resolution grayscale and color cameras and then the ground truths are provided with the help of the laser sensors with which the cars have also been equipped.
Companies making Machine Learning Models for autonomous vehicles
Machine learning models are important in autonomous vehicles for object detection, behavior prediction, and adaptation to changing conditions.
Level 5
Level 5, formerly part of Lyft and now part of Woven Planet (a subsidiary of Toyota), the company is responsible for the development of Toyota's self-driving system. The company's goal is to provide mobility for everyone, whether it be people, goods, or information. In order to do this, they aim to create a self-driving system that is safe and reliable.
comma.ai
OpenPilot is the project that comma.ai is developing a system that can be plugged into most of the cars manufactured in the US and provide them with basic - but incredibly useful - driver assistance features, such as lane keeping or cruise control.
To perform this task, they sell a device that can be plugged into a car - the device itself essentially being a phone with a camera - that is then used to analyze the road imagery and send the information to the car.
Baidu (Apollo)
Baidu is a Chinese multinational technology company specializing in Internet-related services and products. Baidu was established in 2000 and is headquartered in Beijing. The company ranks fourth overall in the Alexa Internet rankings.
Baidu is developing an autonomous vehicle called Apollo. The Apollo project is a driverless vehicle developed by Baidu, Kinglong, and a consortium of more than 40 companies. The Level 4 microcirculation Apolong bus commenced mass production in 2017. It is manufactured by a consortium consisting of Baidu, Kinglong, and SB Drive.
Baidu's Apollo project is aimed at providing a complete solution for autonomous driving, including the hardware, software, and service platforms. The project will open up its technology to other companies and developers to promote the development of autonomous driving.
Companies Making Sensors for Autonomous Vehicles
Sensors are a critical component of autonomous vehicles. They provide the data that the vehicle needs to navigate safely and efficiently. Without sensors, an autonomous vehicle would be blind, unable to see obstacles or properly gauge distances.
So, quite naturally, there are quite a few contenders in the field of producing sensors for AVs.
Continental
Continental is a German company that produces a variety of automotive parts, including autonomous vehicle sensors. The company has a wide array of sensors available, including those for use in cars and trucks.
Continental has a history dating back to 1871, and has been involved in the automotive industry since the early 1900s. In terms of autonomous vehicles, Continental has been working on developing sensors and other technologies that can be used to enable self-driving cars.
Innoviz
Innoviz believes that its innovative products will democratize the field of LIDAR technology and make it available to a variety of industries. The company has developed a solid-state, MEMS LIDAR (Light Detection and Ranging) sensor that is smaller, cheaper, and more reliable than the current LIDAR products on the market.
The company has developed a solid-state, MEMS LIDAR (Light Detection and Ranging) sensor that is more reliable than the current LIDAR products on the market. The company has raised $65 million in funding from investors including Samsung Ventures, SoftBank Ventures Asia, Delphi Ventures, Magma Ventures, Glory Ventures, and Vertex Ventures.
Innoviz is headquartered in Rosh Ha'ayin, Israel, and was founded in 2016 by Omer Keilaf, Oren Rosenzweig, and Amit Steinberg.
Luminar Technologies
Luminar Technologies is a leading provider of light detection and ranging (LiDAR) technology. The Company's mission is to transform mobility with LiDAR-based perception systems, making autonomous vehicles safe and ubiquitous.
Luminar's technology is based on 1550nm silicon photonics and enables a new class of high-performance, cost-effective LiDAR solutions. The Company's technology has been selected by 15 vehicle programs to date.
Luminar has raised over $250 million from strategic investors, including Volvo Cars Tech Fund, as well as leading venture capital firms such as Kleiner Perkins. The Company is headquartered in Orlando, Florida, with offices in Palo Alto, California, and operations in Asia and Europe.
Final Thoughts About The Future Of Autonomous Vehicles
The future of autonomous vehicles is still very much in flux. There are many factors that will determine how quickly technology is adopted and how it impacts our lives.
The key players in the industry are still trying to figure out the best way to bring autonomous vehicles to market. In the meantime, we can expect to see more and more autonomous vehicles on the road.
Recommended Reading
The Many Datasets of Autonomous Driving
Below we'll explore the datasets used to train autonomous driving systems to perform the various tasks required of them.
A System of Record for Autonomous Driving Machine Learning Models
A look at the most useful Weights & Biases features for autonomous vehicle companies
The ML Tasks Of Autonomous Vehicle Development
This report goes through the different tasks in the autonomous vehicle development lifecycle and the various machine learning techniques associated with them.
Object Detection for Autonomous Vehicles (A Step-by-Step Guide)
Digging into object detection and perception for autonomous vehicles using YOLOv5 and Weights & Biases
Add a comment
Tags: Autonomous Vehicles, Beginner
Iterate on AI agents and models faster. Try Weights & Biases today.