Neural network technique for real-time classification automotive problem

Although a large number of neural architectures exist and are applied to a wide range of problems, there continues to be a need for fast real time neural network classifiers, especially in the area of sensor interpretation. Moreover, a need currently exists for cost efficient neural network solutions for automotive applications. An algorithm suitable for this task should be fast and dependable and its hardware platform should be able to operate reliably under challenging conditions such as found in the engine compartment of a vehicle (e.g. temperature, humidity and motion). Work in this area has lead to the idea of neural networks implemented on multiple microprocessor systems (R.J. Howlett and D.H. Lawrence, 1995). The paper describes a novel neural network architecture and implementation, which has the potential to eventually lead to a system that will be able to satisfy the above needs.