A proof-of-concept indirect tire-pressure monitoring system is developed using neural net- works to identify the tire pressure of a vehicle tire. A quarter-car model was developed with Matlab and Simulink to generate simulated accelerometer output data. Simulation data are used to train and evaluate a recurrent neural network with long short-term memory blocks (RNN-LSTM) and a convolutional neural network (CNN) developed in Python with Tensorflow. Bayesian Optimization via SigOpt was used to optimize training and model parameters. The predictive accuracy and training speed of the two models with various parameters are compared. Finally, future work and improvements are discussed.

As the introduction of semiconductor¡Bcomputer¡Bnetwork communcation¡Bmulti-media technology¡K¡Ketc. and the development and application of relevant automotive electronics, the car has becoming a high technological product instead of a traditional and machnical conveyance. Based on IC Insights, over 40% of vehicles will be equipped with automotive electronics and the automotive electronics market will reach the scale of US$192 billion in 2010. The scale of market is expected to reach the record of US$400 billion that represented over 50% of vehicles will be equiped with automotive electronics in 2015. The automotive electronics was divided to six functional field including ¡§Powertrain¡¨¡B¡¨Body¡¨¡B¡¨Chassis¡BSecurity¡¨¡B¡¨Safety¡¨ and ¡§Driver information¡¨. According to the research of IEK, ¡§Tire Pressure Monitoring System¡¨ is the most potential product in ¡§Safety¡¨ field for Taiwan manufacturers of IT industry to invest their resource. Based on Strategy Analytics, the global market of Tire Pressure Monitoring System reaches the scale of US$200 million in 2004 and the growth is expected to reach the record of US$1,279 million in 2008 and 63.6% annual compound growth. Quality Function Deployment (QFD) is a systematical tool with customer orientated concept. Adopting QFD method to product design efficiently is able to shorten development time¡Breduce cost and improve quality to satisfy customer¡¦s needs in his or her mind. This research is completed by case study of Tire Pressure Monitoring System (TPMS) which is developed by certain Taiwan company with potentials that is chosen by specialists. In this research we try to collect related information about certain company and explore customer¡¦s real-life needs by conducting questionnaire surveys. Kano¡¦s two-dimensional quality model is applied to identify customer¡¦s critical quality requirements. Quality factor¡Bsub-system/component and process are deployed by the matrix method of QFD to recognize key items for improving product design.