Diagnostic Analysis for an Autonomous Truck Using Multiple Attribute Decision Making

Autonomous vehicles are robots capable of navigate in urban or confined spaces with no human intervention. For that, the vehicle is equipped with specific hardware and software components. To accomplish the task of navigate in urban scenarios, some aspects must be analyzed, including the system integrity. In addition, the robot must make decisions based on sensor information. This paper proposes the use of Multiple Attributes Decision Making and Fuzzy sets, for an automatic analyzer of diagnostic information of an Autonomous Truck, and its perception data. The decision-maker task was to choose a safe maneuver to avoid or overcome a dangerous situation. Thus, Analytic Hierarchical Process (AHP) was used to compute the weights of the attributes, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to rank the alternatives, and Fuzzy to model the decision matrix. In the end, the system was tested in a real scenario where different events were created to validate the performance. The results show consistency in choices, which indicates the suitability of the approach for the task.