Dataset of the intermediate competition in challenge MALIN: Indoor–outdoor inertial navigation system data for pedestrian and vehicle with high accuracy references in a context of firefighter scenario

Abstract This paper provides a multiple sensor dataset collected by the CyborgLOC team during intermediate competition of Challenge MALIN (MAitrise de la localisation INdoor), which is for indoor/outdoor real-time positioning. The sensors, including GNSS receiver Ublox NEO-M8N, Realsense D435i stereo camera, three Xsens MTi-300 and one PERSY (PEdestrian Reference SYstem), are mounted on different parts subject's body. foot-mounted positioning device with tri-axial accelerometer, gyroscope, magnetometer as well M8T. two scenarios designed in training centre firefighters CFIS (Fire Rescue Training Centre) Blois, France to simulate situation interventions. With total distances around 2 km each scenario, travelled trajectories passed through challenging environments indoor, outdoor, urban canyon. indoor part contains stair levels, from underground up 6th floor. travel modes vehicles pedestrians. Several classical activities realized such walking, running, stair-climbing, side-walking, crawling, passing above/below obstacles, carrying stretcher, ladder climbing, etc. High accurate ground truth stationary points enclosing volumes provided organizers competition, i.e., Directorate General Armaments (DGA: Direction Generale l'Armement). Provided raw data, they allow evaluation performances. available data repository https://doi.org/10.5281/zenodo.4290789 .