Three-Year Review of the 2018–2020 SHL Challenge on Transportation and Locomotion Mode Recognition From Mobile Sensors

The Sussex-Huawei Locomotion-Transportation (SHL) Recognition Challenges aim to advance and capture the state-of-the-art in locomotion transportation mode recognition from smartphone motion (inertial) sensors. goal of this series machine learning data science challenges was recognize eight activities (Still, Walk, Run, Bus, Car, Train, Subway). three focused on time-independent (SHL 2018), position-independent 2019) user-independent 2020) evaluations, respectively. Overall, we received 48 submissions (out 93 teams who registered interest) involving 201 scientists over years. survey captures through a meta-analysis contributions challenges, including approaches, performance, computational requirements, software tools frameworks used. It shown that methods can distinguish with relative ease most modes transportation, although differentiating between subtly distinct activities, such as rail transport (Train Subway) road (Bus Car) still remains challenging. We summarize insightful participants could be employed address practical recognition, for instance, tackle over-fitting, employ robust representations, exploit augmentation, smart post-processing techniques improve performance. Finally, present baseline results compare unified pipeline decision window length.