The Backseat Driver – Driver Advisory System 2019-01-0880
Today, most of the V2X, ADAS and autonomous driving systems are based on precise location and prediction of movement. These systems are computationally complex and depend on precise sensor measurements. This might not be always possible e.g. inaccurate GPS location during cloudy weather condition.
Proposed is an approach which is based on heuristics and machine learning concepts to model driving guidance. This is very similar to how “humans” drive. Our proposed approach complements the existing V2X and RADAR based systems by issuing an advisory to the driver. A machine learning approach (Artificial Neural Networks, Expectation-Maximization, Decision trees) can be adapted to generate advisories. Moreover, with continuous reinforced learning, the predictions will become more accurate. In the upcoming days of semi-autonomous driving, heuristic approach for predicting and issuing advisories can make a difference to the driving experience.
The Backseat Driver uses historical data and machine leaning approach to generate advisories. Timely advisories such as merging traffic, approaching traffic when taking a turn, busy pedestrian crossing, school/hospital zone could alert driver to be cautious.
The Backseat Driver is a low complexity, low cost system which works even if sensor data is not accurate. Unlike V2X, there is no critical number of installations required for the system to work. In contrast to V2X and ADAS systems, which provide advisories only on approaching a situation, the Backseat Driver can issue advisories in advance.
The current scope of research shall be limited to issuing an advisory to the driver and not to control any functions within a vehicle.
In conclusion, the research focus is to provide real time and in advance advisory messages to the driver, without dependency on availability of accurate sensor data.