The research in the complexity of environment and driving task plays a significant role in the test and validation of L3+ autonomous vehicles (AVs). We define the complexity of environment and driving task of AVs as the sum of uncertainties in the prediction process and propose a method to quantify it. First, based on the analysis of the effect factors of perception system and the data processing algorithm of the cognition system of AV, as well as combining with complexity theory and the definition mentioned above, several types of environmental complexity factors liked the amount, variety and path of motion of objects, the environment conditions are abstracted. Next, multi-factors analysis quantification method is used to generate a quantifying model for the complexity of environment. The driving task of AV defines which environment complexity needs to be quantified. Therefore, the quantification result of the complexity of environment and driving task is the accumulation of the environmental complexities. Finally, a survey among 5 experts in the field of autonomous driving is conducted. The consistency between the average survey results and the quantification by using this methodology is 81%, which proves the effectiveness of this methodology.

Autonomous Vehicles Testing; Complexity of Environment; Driving Task;