**7. Conclusions**

Technical advancements and the ongoing developments in sensor technology and data science promise to unlock huge potentials for the diagnosis and understanding of autism, and for supporting affected people with training or intervention programs that can be tailored to their specific needs. At the same time, living up to these potentials calls for a concerted and interdisciplinary effort in which computer scientists, engineers, psychologists, and neuroscientists jointly collaborate in large-scale research projects that can uncover, in a quantitative manner, the efficiency of these approaches. In our view, this will be the route not only for establishing routine contributions to evidence-based diagnosis and interventions in autism [89] but also to ensure that more people with autism can genuinely benefit from tailor-made technology.

**Funding:** Previous research by SRS on related topics has been funded by a grant from the Bundesministerium für Bildung und Forschung (BMBF), in a project on an irritation-free and emotion-sensitive training system (IRESTRA; Grant Reference: 16SV7210), and another BMBF project on the psychological measurement of anxiety in human-robot interaction (3DimIR, Grant Reference 03ZZ0459B).

**Acknowledgments:** AEK and SRS would like to thank the Herbert Feuchte Stiftungsverbund for supporting the research in the Social Potentials in Autism Research Unit (www.autismus.uni-jena.de). SRS would like to thank the Swiss Center for Affective Sciences at the University of Geneva, Switzerland, for hosting a sabbatical leave in summer 2019 during which this paper was written.

**Conflicts of Interest:** The authors declare no conflict of interest. In particular, funding bodies had no role in the planning, collection, or interpretation of evidence reviewed in this paper; in the writing of the manuscript, or in the decision to publish the results.
