• Technology collaboration to develop high-performance intelligent driving solutions designed to simulate human behaviors for more natural traffic understanding and driving with neural networks

• The collaboration will optimize end-to-end Birds Eye View and transformer models on Snapdragon Ride Platforms, improving performance in complex scenarios such as human-vehicle interaction, detours, and narrow road driving

April 8th, 2025 — DeepRoute.ai, a leading global artificial intelligence company, and Qualcomm Technologies, Inc., announced today a technology cooperation to develop multiple high performance and cost-effective advanced intelligent driving solutions for advanced driver assistance systems and automated driving (ADAS/AD) based on Snapdragon RideTM Platforms (SA8650). 

DeepRoute.ai and Qualcomm Technologies, Inc. signed cooperation agreement

Utilizing a Snapdragon Ride Platform, DeepRoute.ai’s intelligent driving solutions will cover LiDAR and vision-only ADAS systems, both of which support advanced intelligent driving features including Urban NOA, Highway NOA and automated parking. The solution can be deployed on fuel vehicles and new energy vehicles, meeting the personalized needs of global automakers for a wide range of vehicles, meeting the personalized needs of global automakers for a wide range of vehicles Through the technology collaboration, the companies will also work together to optimize AI models, such as end-to-end (E2E) Birds Eye View (BEV) and transformers, on Snapdragon Ride Platforms.

The architecture simulates human neural network, which enables the intelligent driving systems to understand the traffic environment and the logic behind driving behaviors more naturally. Especially in long-tail scenarios such as human-vehicle negotiation, nudging around parked vehicles, passing irregular intersections and narrow road driving, the intelligent driving system can achieve better performance in complex road scenarios. The end-to end model can help the intelligent driving system understand human’s driving culture. For instance, when the road ahead is blocked due to construction, vehicles equipped with the (E2E) model can analyze the traffic conditions in real time and promptly change lanes to support efficient passage.

For safety, DeepRoute.ai has optimized its algorithms for complex lighting conditions such as strong light, low light, and backlight. By enhancing the perceptual limitations of vision-only systems in extreme environments, it delivers core performance on par with LiDAR-based solutions.

This cooperation has leveraged the technical advantages of the Snapdragon Ride Platforms. Through an open architecture, the platforms allow automakers and Tier-1 suppliers to deploy various ADAS algorithms including camera perception, sensor fusion, driving strategies, automated parking and driver monitoring. Integrated with AI accelerator and image processing engine, the platforms can concurrently process data from 16 cameras, multiple radars and lidars to enable real-time environment perception and decision making like pedestrian detection, lane detection and obstacle avoidance, aimed at fully meeting the needs of advanced intelligent driving solutions.

“The Snapdragon Ride Platforms from Qualcomm Technologies, Inc. have been one of the preferred choices for global automakers,” said Maxwell Zhou, CEO of DeepRoute.ai. “I believe the advanced intelligent driving solutions built upon Snapdragon Ride Platforms will bring even better performance to advanced intelligent driving features like Urban NOA and help accelerate the widespread adoption of intelligent driving.”

“Our technology cooperation with DeepRoute.ai is a collaborative effort that harnesses DeepRoute.ai’s algorithmic expertise and the technological advantages of working utilizing Snapdragon Ride Platforms,” said Anshuman Saxena, Vice President, Product Management, Qualcomm Technologies, Inc. “Together we aim to reduce hardware costs and deliver advanced intelligent driving experiences to users globally. We look forward to continuing our relationship with DeepRoute.ai and their proven expertise in integrating end-to-end AI models into mass-produced vehicles.”