In this work we implement and build a prototype of a new ETA device to help the walking of blind people. The ETA, based on two kind of sensors, is able to intercept positive and negative obstacles and to warn the user on time. Characterizations of the sensors are made to understand their main features and the logic to take the decision via software. Moreover it is studied the geometry of the system human-device to verify if the distance range of the infrared sensor is respected and to estimate the time of the interception of the obstacle. The goal of this system is the hybrid configuration of the sensors that is used for the first time in this kind of application,this able us to take the best properties of the two sensors making the hole system precise and reliable. Moreover we conduce an energy consumption analysis in order to highlight the importance of the autonomy of these devices. As far as the authors know this is one of the first work in which the ETA is tested walking, introducing another noise, due to the human oscillation, on the main noise of the sensors. In order to clean the input signals and make the system usable we implement simple filters based on Butterworth, Kalman and moving average approach. Future work will focus the attention on using better filters to make the device faster in the interception of the obstacle i.e. to reduce the delay time and the amplitude of the noise.