LOGO

BMW i Ventures Invests in Kodiak Robotics - Autonomous Trucking

June 24, 2021
BMW i Ventures Invests in Kodiak Robotics - Autonomous Trucking

BMW i Ventures Invests in Autonomous Trucking Firm Kodiak Robotics

BMW’s venture capital division, based in Silicon Valley, is providing funding to Kodiak Robotics, a developer of self-governing trucking technologies.

The received capital will be allocated towards strengthening the safety validation of Kodiak’s autonomous system, accelerating its path to commercial deployment. Expansion of the vehicle fleet and recruitment of skilled personnel are also key objectives, with a target of at least doubling the current operational vehicle count annually.

Deal Details and Strategic Alignment

The financial specifics of the investment remain undisclosed. While BMW i Ventures typically focuses on companies offering solutions relevant to BMW’s business, Kodiak’s CEO and co-founder, Don Burnette, clarified to TechCrunch that this investment is purely financial, lacking a direct technical collaboration between the two entities.

This funding round follows a recent minority stake acquisition in Kodiak by tire manufacturer Bridgestone, the details of which were also not publicly revealed. To date, Kodiak has announced $40 million in Series A funding, with Burnette indicating several additional investments have been secured since then.

Cost-Effective Driverless Operations

Kodiak aims to achieve fully driverless operations at a significantly lower cost than competitors. Burnette stated the company intends to reach this milestone for less than 10% of the $5.7 billion raised by Waymo, and under 25% of TuSimple’s $1.94 billion total funding – approximately $500 million.

“Our efficiency, both historically and going forward, will allow us to achieve driverless capabilities with a considerably smaller financial outlay,” Burnette explained.

The investment from BMW i Ventures will contribute to Kodiak’s Series B funding round. The focus isn’t on enhancing the self-driving capabilities themselves, but rather on establishing a robust safety case and demonstrating the system’s ability to operate safely without a driver.

Autonomy Levels and Current Capabilities

“We are progressing towards Level 4 autonomy, while maintaining a driver for continuous system monitoring,” Burnette stated. “Currently, we operate as a Level 2 system, a classification shared by most in the industry.”

Level 2 autonomy, as defined by the Society of Automobile Engineers (SAE), involves driver assistance features like lane centering and adaptive cruise control. Level 4 signifies the vehicle’s capacity to manage all driving tasks under specific conditions without human intervention.

Kodiak reports advancements in this area, having recently completed “disengagement-free deliveries” on a commercial route between Dallas and Houston in January. This route, operational for two years, has reached a level of maturity where the system can effectively handle highway conditions.

Sensor Fusion and Real-Time Perception

The company is adept at complex maneuvers, including identifying and responding to vehicles stopped on the roadside. The system automatically slows down or adjusts its lane to provide additional clearance, mirroring human driving behavior.

To validate the safety of its technology, Kodiak is accumulating data through simulations, closed-track testing, and real-world driving. Burnette emphasizes that Kodiak uniquely avoids prioritizing a single sensor type, adopting a comprehensive approach instead.

“We recognize the strengths and weaknesses of each sensor modality and aim to leverage the former while mitigating the latter through sensor fusion,” Burnette said. “Our algorithm intelligently selects the most advantageous sensors based on the prevailing conditions.”

Data Collection and Team Expertise

Kodiak does not rely on high-definition (HD) mapping, enabling its trucks to perceive the environment in real-time and adapt to changing road conditions. Data used for training the system is gathered from Kodiak’s fleet and engineered scenarios, with auto-labeling facilitated by Scale AI, contributing to cost efficiency.

The Kodiak team comprises veterans from Google’s original self-driving initiative, Uber, Lyft, and other prominent tech firms. BMW i Ventures’ investment followed a rigorous evaluation process, including scrutiny by their autonomous driving specialists.

#BMW i Ventures#Kodiak Robotics#autonomous trucks#autonomous driving#trucking technology#investment