Search Results

Electronic braking system (EBS) of commercial vehicle is developed from ABS to enhance the brake performance. Based on the early development of controller hardware, this paper starts with an analysis of the definition of EBS. It aims at the software design of electronic control unit, and makes it compiled into the controller in the form of C language by the in-depth study about control strategy of EBS in different braking conditions. Designed controller software is divided into two layers. The upper control strategy includes the recognition algorithm of driver's braking intention, estimation algorithm of the vehicle state, conventional braking strategy which consists of the algorithm of deceleration control and braking force distribution, and emergency braking strategy which consists of the algorithm of brake assist control and ABS control.

Pneumatic Electric Braking System (EBS) is getting widely spread for commercial vehicles. Pneumatic EBS improves the problem of slow response of traditional pneumatic braking system by implementing brake-by-wire. However, the time-delay response and hysteresis of some electro-pneumatic components and some other issues decrease the response and control accuracy of the pneumatic EBS.

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.