Autonomous Vehicle Startup Phantom AI Raises $22 Million in Series A Funding for its Advanced Driver Assistance System

Autonomous Vehicle Startup Phantom AI Raises $22 Million in Series A Funding for its Advanced Driver Assistance System

Author: Eric Walz   

Autonomous vehicle startup Phantom AI announced it has raised $22 million in Series A financing led by Celeres Investments and joined by U.S. automaker Ford Motor Co. and KT,  South Korea's largest telecommunications company. Existing investors Millennium Technology Value Partners and DSC Investment also participated in the round. 

Phantom AI previously raised $5 million in seed funding, bringing the total capital raised to date to $27 million. The Burlingame, California company said it will use the new financing to accelerate product development while scaling its operations in Europe and Asia.

Phantom AI was founded in 2016 by lead engineers Hyunggi Cho and Chan Kyu Lee from Tesla Motors and Hyundai respectively, where they developed ADAS (Advanced Driver Assistance Systems) for the two automakers. Cho, who co-founded Phantom AI and serves as the company's CEO, was a perception engineer at Tesla, contributing to the Autopilot system used in the Tesla Model S.  

"To the automakers frustrated with the lack of options in computer vision technologies—Phantom AI is here to help. We are thrilled to bring our AI-based perception technology, including computer vision, sensor fusion and control capabilities to market, and to have the support of our new investors to help us accelerate production globally," said Cho.

Advanced driver assist systems (ADAS) include newer vehicle safety technologies, such as emergency braking, or adaptive cruise control that maintains the ideal highway speed while keeping a safe distance from other vehicles. 

Tesla's Autopilot automated driving feature is an example of an ADAS. It can perform functions such as keeping the vehicle centered in a highway lane by determining the vehicle's precise position from data collected from a forward-facing camera and controlling the steering, for example.

In some newer vehicles, the ADAS allows some level of automated driving in certain situations. The next step for ADAS applications is developing cars capable of self-driving without human intervention. However, these future self-driving vehicles will use the same perception and computer vision technologies being developed by Phantom AI.

Phantom AI's customers are Tier 1 automotive manufacturers working on level 2-3 autonomous driving solutions. The company is working with several unnamed automakers, a truck OEM and multiple European and Asian tier 1 suppliers, offering a robust yet low cost solution so automakers can add advanced ADAS to their vehicles at scale.

Phantom AI Developed Three Core Products for ADAS

Phantom AI has three core products, PhantomVision, PhantomFusion and PhantomDrive.

PhantomVision is a deep learning based computer vision system that acts as the "eyes" of a vehicle. It's a visual perception engine which uses single or multiple 360 degree cameras to automatically recognize road objects in the path of a vehicle, such as other vehicles or pedestrians. It can also be used in autonomous vehicles as part of a vehicle's ADAS.

The PhantomVision perception engine can track and identify other vehicles, pedestrians, bicyclists, traffic signs and traffic lights. The multi-camera system covers a 360 degree view from the vehicle without any blind spots.

The real-time detection and target tracking of PhantomVision supports various ADAS systems, including AEB (Auto Emergency Braking), ACC (Adaptive Cruise Control), TJA (Traffic Jam Assist) and LKAS (Lane Keeping Assist System). 

PhantomVision is a low cost, scalable solution for automakers, since it can be deployed on low power embedded SoCs (System on Chip) using hardware accelerators to run deep learning algorithms and robust computer vision processing.

Phantom AI's PhantomFusion product is an independent sensor fusion and object tracking system. It fuses all of the raw data collected from various vehicle sensors like cameras, radar and LiDAR. It can fuse data from up to 6 cameras and 6 radars, 360 degrees around a vehicle. 

PhantomFusion allows for faster integration of a perception system and more uniform tracking in the areas needed for various levels of autonomy, according to Phantom AI.

The company's third product is PhantomDrive. It's a vehicle controller that predicts the motion of objects once they are detected and identified. The program understands road and traffic from perception systems, but it goes a step further by screening out any false-positive information and compensates for any missing information by rational reasoning. 

The system assesses any collision risk with other objects and decides appropriate vehicle actions based on risk assessment. For urban driving, knowledge of a vehicle's precise position is a critical piece of information for ADAS or for self-driving cars to execute safety-related, autonomous driving maneuvers, such as a sudden lane change. 

PhantomDrive determines a vehicle's trajectory, then controls acceleration, braking and steering. It can also handle any emergency situation such as stopping for a pedestrian in the roadway. 

PhantomDrive also supports typical driver safety features such as Adaptive Cruise Control, Automatic Lane Keeping and Automatic Lane Change.

"With over 1.3 million fatalities resulting from traffic accidents every year, we believe a lot more can be done with computer vision and ADAS technologies to help protect drivers, pedestrians and cyclists and property," said Ray Cheng, partner at Millennium, an investment firm that has also backed Aurora Innovation. Aurora co-founder Chris Urmson once led Google's self-driving car project, which is now known as Waymo.

Analysts are predicting the ADAS market will reach $135 billion by 2027, up from $30 billion in 2019. Consulting firm McKinsey anticipates that demand for ADAS features that support automatic braking and steering will increase through 2026, spurred by consumer and regulatory interest in vehicle safety systems that reduce accidents and save lives. 

Some of these new vehicle safety features are already mandated. New regulations in Europe require vehicles produced after 2014 to be equipped with Automatic Emergency Braking (AEB) systems and forward collision warning systems. In the U.S., 20 automakers pledged to voluntarily equip all new passenger vehicles with AEB by September 1, 2022, in order to help prevent front-to-rear crashes.  

Phantom AI develops ADAS features for cars at all price points, Phantom AI's products are targeted at automakers and Tier 1 suppliers that are looking to add, scalable, Level 1 level 3 autonomous driving capabilities to their new vehicles.

Eric Walz
Eric Walz
Originally hailing from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry here in Silicon Valley. He has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology.
Prev:Toyota Patents Autonomous Workhorse That Can Help You Shop Next:Tesla Engineers Share Details of a Prototype Ventilator Made With Electric Car Parts
    view more