Design of a multimodal device to improve well-being of autistic workers interacting with collaborative robots

Several studies report that people diagnosed with autism spectrum disorder (ASD) have low employment rates and major difficulties in maintaining occupation (Hendricks, 2010). Adopting smart and interactive technology, promoted by Industry 4.0, can be a valid solution to improve social inclusion in the workplace. The present work describes the design and development process of the multimodal device (A)MICO (acronym for 'A Multimodal device to improve inclusive Interaction between Cobot and Operator'), aimed to improve the user experience of ASD workers interacting with collaborative robots (cobots) in production lines. (A)MICO proposes a new intuitive mode of communication in which information about the cobot activity is transferred through acoustic and visual feedback. Lights and sounds are designed to make the user aware of the cobot's activity and gain control of the situation. To define the optimal feedback configuration, a co-design process has been performed involving five users with high functioning autism (HFA). The inclusion of users with HFA in the design process allows to analyse the system from different perspectives and increase the acceptance of the final product (Francis et al., 2009). In addition, (A)MICO is designed to be integrated into a pre-existing cobot system and its feedback can be customized making job tasks more human-friendly and accessible to all. To achieve this purpose, Design for All principles were taken into consideration since the preliminary stages of the study, in order to develop a device that could represent a valuable aid in improving the well-being and productivity for both neurotypical and neurodiverse users.

Experience in Engineering Complex Systems: Active Preference Learning with Multiple Outcomes and Certainty Levels

Black-box optimization refers to the optimization problem whose objective function and/or constraint sets are either unknown, inaccessible, or non-existent. In many applications, especially with the involvement of humans, the only way to access the optimization problem is through performing physical experiments with the available outcomes being the preference of one candidate with respect to one or many others. Accordingly, the algorithm so-called Active Preference Learning has been developed to exploit this specific information in constructing a surrogate function based on standard radial basis functions, and then forming an easy-to-solve acquisition function which repetitively suggests new decision vectors to search for the optimal solution. Based on this idea, our approach aims to extend the algorithm in such a way that can exploit further information effectively, which can be obtained in reality such as: 5-point Likert type scale for the outcomes of the preference query (i.e., the preference can be described in not only "this is better than that" but also "this is much better than that" level), or multiple outcomes for a single preference query with possible additive information on how certain the outcomes are. The validation of the proposed algorithm is done through some standard benchmark functions, showing a promising improvement with respect to the state-of-the-art algorithm.