Search Results

An autonomous vehicle is a comprehensive intelligent system that includes environment sensing, vehicle localization, path planning and decision-making control, of which environment sensing technology is a prerequisite for realizing autonomous driving. In the early days, vehicles sensed the surrounding environment through sensors such as cameras, radar, and lidar. With the development of 5G technology and the Vehicle-to-everything (V2X), other information from the roadside can also be received by vehicles. Such as traffic jam ahead, construction road occupation, school area, current traffic density, crowd density, etc. Such information can help the autonomous driving system understand the current driving environment more clearly. Vehicles are no longer limited to areas that can be sensed by sensors. Vehicles with different autonomous driving levels have different adaptability to the environment.

It’s hard to achieve complete self-driving using hand-crafting generalized decision-making rules, while the end-to-end self-driving system is low in complexity, does not require hand-crafting rules, and can deal with complex situations. Modular-based self-driving systems require multi-task fusion and high-precision maps, resulting in high system complexity and increased costs. In end-to-end self-driving, we usually only use camera to obtain scene status information, so image processing is very important. Numerous deep learning applications benefit from multi-task learning, as the multi-task learning can accelerate model training and improve accuracy with combine all tasks into one model, which reduces the amount of calculation and allows these systems to run in real-time. Therefore, the approach of obtaining rich scene state information based on multi-task learning is very attractive. In this paper, we propose an approach to multi-task learning for semantics, geometry and motion.