Commonsense Visual Sensemaking For Autonomous Driving On Generalised Neurosymbolic Online Abduction Integrating Vision And Semantics Jakob Suchan Mehul Bhatt Srikrishna Varadarajan by Jakob Suchan & Mehul Bhatt & Srikrishna Varadarajan instant download after payment.

We demonstrate the need and potential of systematically integrated vision and semanticssolutions for visual sensemaking in the backdrop of autonomous driving. A generalneurosymbolic method for online visual sensemaking using answer set programming(ASP) is systematically formalised and fully implemented. The method integrates stateof the art in visual computing, and is developed as a modular framework that isKeywords:generally usable within hybrid architectures for realtime perception and control. Weevaluate and demonstrate with community established benchmarks KITTIMOD, MOTCognitive vision2017, and MOT-2020. As use-case, we focus on the significance of human-centredDeep semanticsDeclarative spatial reasoningvisual sensemaking —e.g., involving semantic representation and explainability, questionKnowledge representation and reasoninganswering, commonsense interpolation— in safety-critical autonomous driving situations.Commonsense reasoningThe developed neurosymbolic framework is domain-independent, with the case ofVisual abductionautonomous driving designed to serve as an exemplar for online visual sensemakingAnswer set programmingin diverse cognitive interaction settings in the backdrop of select human-centred AIAutonomous drivingtechnology design considerations.Human-centred computing and design© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under theStandardisation in driving technologyCC BY-NC-ND license ( cognition and AI