A Search-Based Framework for Automatic Generation of Testing Environments for Cyber-Physical Systems

Abstract

Background. Many modern cyber-physical systems incorporate computer vision technologies, complex sensors and advanced control software, allowing them to interact with the environment autonomously. Examples include drone swarms, self-driving vehicles, autonomous robots, etc. Testing such systems poses numerous challenges: not only should the system inputs be varied, but also the surrounding environment should be accounted for. A number of tools have been developed to test the system model for the possible inputs falsifying its requirements. However, they are not directly applicable to autonomous cyber-physical systems, as the inputs to their models are generated while operating in a virtual environment. Aims. In this paper, we aim to design a search-based framework, named AmbieGen, for generating diverse fault-revealing test scenarios for autonomous cyber-physical systems. The scenarios represent an environment in which an autonomous agent operates. The framework should be applicable to generating different types of environments. Method. To generate the test scenarios, we leverage the NSGA-II algorithm with two objectives. The first objective evaluates the deviation of the observed system's behaviour from its expected behaviour. The second objective is the test case diversity, calculated as a Jaccard distance with a reference test case. To guide the first objective we are using a simplified system model rather than the full model. The full model is used to run the system in the simulation environment and can take substantial time to execute (several minutes for one scenario). The simplified system model is derived from the full model and can be used to get an approximation of the results obtained from the full model without running the simulation. Results. We evaluate AmbieGen on three scenario generation case studies, namely a smart-thermostat, a robot obstacle avoidance system, and a vehicle lane-keeping assist system. For all the case studies, our approach outperforms the available baselines in fault revealing and several other metrics such as the diversity of the revealed faults and the proportion of valid test scenarios. Conclusions. AmbieGen could find scenarios, revealing failures for all the three autonomous agents considered in our case studies. We compared three configurations of AmbieGen: based on a single objective genetic algorithm, multi-objective, and random search. Both single and multi objective configurations outperform the random search. Multi objective configuration can find the individuals of the same quality as the single objective, producing more unique test scenarios in the same time budget. Our framework can be used to generate virtual environments of different types and complexity and reveal the system's faults early in the design stage.