Robust Data Fusion of Multimodal Sensory Information for Mobile Robots

Abstract

Urban search and rescue (USAR) missions for mobile robots require reliable state estimation systems resilient to conditions given by the dynamically changing environment. We design and evaluate a data fusion system for localization of a mobile skid‐steer robot intended for USAR missions. We exploit a rich sensor suite including both proprioceptive (inertial measurement unit and tracks odometry) and exteroceptive sensors (omnidirectional camera and rotating laser rangefinder). To cope with the specificities of each sensing modality (such as significantly differing sampling frequencies), we introduce a novel fusion scheme based on an extended Kalman filter for six degree of freedom orientation and position estimation. We demonstrate the performance on field tests of more than 4.4 km driven under standard USAR conditions. Part of our datasets include ground truth positioning, indoor with a Vicon motion capture system and outdoor with a Leica theodolite tracker. The overall median accuracy of localization—achieved by combining all four modalities—was 1.2% and 1.4% of the total distance traveled for indoor and outdoor environments, respectively. To identify the true limits of the proposed data fusion, we propose and employ a novel experimental evaluation procedure based on failure case scenarios. In this way, we address the common issues such as slippage, reduced camera field of view, and limited laser rangefinder range, together with moving obstacles spoiling the metric map. We believe such a characterization of the failure cases is a first step toward identifying the behavior of state estimation under such conditions. We release all our datasets to the robotics community for possible benchmarking.

A Review of Point Cloud Registration Algorithms for Mobile Robotics

“…The result of this estimation can then be taken as the initial transformation. But the result of the registration can also be integrated in a filtering approach, for instance as an observation of an extended Kalman filter [Kubelka et al, 2014].…”

Non-uniform sampling strategies for continuous correction based trajectory estimation

“…As high uncertainty is associated to encoder-based odometry of such skid-steering vehicles, the encoders and IMU measurements are pre-fused with the technique presented in Kubelka et al [10]. The resulting odometry estimate is the one to be processed by Algorithm 1.…”

Relaxing the planar assumption: 3D state estimation for an autonomous surface vessel

“…The registration of subsequent point clouds provides relative position updates and, when going a step further, the comparison against a known map links the position information to a global frame. Rectifying laser scan measurements via state prediction during the swiping has been done for other types of robots (e.g., ground robots [Kubelka et al, 2014] and quadrotors [Michael et al, 2012]). Another approach is to include the laser information directly in the state estimation [Bosse and Zlot, 2009].…”

“…In robotics, rotating 2D laser rangefinders are often used to provide similar 3D point clouds at considerably lower price (e.g., [Scherer et al, 2012, Kubelka et al, 2014). In fact, for the shoreline mapping application which we describe below, we also used a nodding 2D laser rangefinder (see Figure 6).…”