Smartphone-Based Indoor Floor Plan Construction via Acoustic Ranging and Inertial Tracking

The lack of indoor floor plans is one the major obstacles to ubiquitous location-based services. Dedicated mobile robots with high-precision sensors can measure and produce accurate maps, but deployment remains low for public. Computer vision techniques are adopted by some existing smartphone-based methods build 3D point cloud, which have cost a quantity efforts image collection risk privacy issues. In this paper, we propose BatMapper-Plus adopt acoustic ranging inertial tracking construct precise complete on smartphones. It emits signals distance from smartphone neighbouring wall segment, produces accessible areas surrounding building during walking. also refines constructed plan eliminate scattered segments, identifies connection areas, including stairs elevators among different floors. addition, an LSTM-based dead-reckoning model trained outdoor IMU readings GPS records, use it infer step length walking, thereby improving quality. We elaborate how map navigation, i.e., Dynamic Time Warping algorithm automatically matches current historical sensory data construction fine-grained walking guidance. To show our effectiveness compared state-of-the-art, carry out extensive experiments in teaching residential building. proves that method efficient without any concerns texture/illumination limitations.