Monte Carlo Localization in Dense Multipath Environments Using UWB Ranging

Online model estimation of ultra-wideband TDOA measurements for mobile robot localization

“…The robot's selflocalization algorithm is based on UWB measurements, yet it does not employ an UWB error model, and instead relies on a least squares method to solve the multilateration problem. The studies in [6] and [7] develop probabilistic models for biased UWB range measurements which are combined with on-board odometry data. Yet, both papers model NLOS biases within augmented-state particle filters that do not take LOS/NLOS signal path conditions and bias probability distributions into account explicitly, and might therefore be limited by this simplified approach.…”

WLAN-Based Pedestrian Tracking Using Particle Filters and Low-Cost MEMS Sensors

“…Different from other expensive IMU assisted systems [23,24], our framework integrates the typical WLAN pedestrian positioning system with only a low cost accelerometer and map information, as shown in Fig. 1.…”

Cooperative Tasks between Humans and Robots in Industrial Environments

“…The fast measurements of the inertial motion capture system can be easily represented with the linear Gaussian models of a Kalman filter, while the slow measurements of the UWB system are better modelled with a non-linear nonGaussian technique (such as a particle filter) because of their high latency. The use of the efficient Kalman algorithm for inertial measurements improves the computational performance of the fusion algorithm in comparison with other fusion algorithms, which only apply particle filters to all measurements [26][27][28]. The use of the particle filter in order to correct the inertial error when UWB measurements are received improves the precision of the position estimates in comparison with previous fusion approaches, which only apply Kalman filters to all measurements [29][30][31].…”