Search papers, labs, and topics across Lattice.
This paper introduces EVITA, a novel approach that employs multi-objective optimization to generate diverse and complex driving scenarios for testing interacting autonomous vehicles (AVs). By addressing the combinatorial explosion of interactions among multiple AVs, EVITA effectively uncovers safety-critical behaviors that traditional single-AV testing methods overlook. Experimental results demonstrate that EVITA significantly increases the variety of AV interactions compared to state-of-the-art techniques, enhancing the overall safety assessment of autonomous systems.
EVITA reveals safety-critical interactions among multiple autonomous vehicles that traditional testing methods fail to capture.
Autonomous vehicles (AVs) must be thoroughly tested to meet high safety standards and avoid endangering both AV passengers and road users. Scenario-based testing implements driving scenarios in virtual simulation environments as a cost-effective alternative to field testing. Common scenario-based testing approaches set the environment and the surrounding traffic and test a single AV. Recent studies show that the approaches that test single AVs miss critical behaviors that emerge from interactions among multiple AVs. Effective approaches to test scenarios that emerge from n-way interactions must address the combinatorial explosion that the presence of multiple AVs further exacerbates. In this paper, we propose EVITA, an approach that leverages multi-objective optimization to generate scenarios that trigger multiple and diverse AVs interactions, while minimizing the complexity of the generated scenarios, to effectively test multiple interacting AVs and reveal safety-critical scenarios that current approaches overlook. The experimental results that we discuss in this paper confirm that EVITA triggers a higher variety of AVs interactions than state-of-the-art approaches, thus improving the likelihood to reveal safety-critical behaviors.
