IROS 2016 State Estimation and Terrain Perception

Enabling mobile robots to move in unknown environments requires them to have an understanding of the surrounding and their location within. This is especially important for unmanned ground vehicles (UGVs) such as legged robots, tracked and wheeled vehicles. These systems rely on the interaction with the terrain for traction, while avoiding collision, slippage, and instabilities. Therefore, an accurate terrain representation together with a reliable pose estimation is crucial for the successful deployment of these robots. The main objective of this workshop is to bring together people from the fields of state estimation and terrain perception for UGVs to exchange ideas and foster collaboration.

Since in real world robotic applications both state estimation and terrain perception are tightly interweaved problems, they have to be addressed conjointly. In comparison to previous similar workshops, the proposed workshop pays special attention on the challenges resulting from combining methods from both state estimation and terrain perception. The discussed topics will include the processing of exteroceptive sensor data (vision, RGB-D, LiDAR, Time-of-Flight), the representation and interpretation of environments, and the consistent formulation of state estimation algorithms. During the discussions we will also address the practical problems related to the integration on real robots where uncertain, corrupted and missing data have to be handled.

Important dates

• Poster abstract submission deadline (extended): August 26, 2016
• Notification of acceptance: August 31, 2016
• Workshop: October 14, 2016

Topics of interest

The key topics of interest for this workshop include:

• Onboard sensing for 3D mapping of unstructured environments
• Modelling and handling of uncertainties, drift, and outliers
• Terrain representation and estimation
• Tight integration of pose estimation and mapping
• Integration of dynamical models and kinematic constraints
• Calibration and modeling of sensors
• Integration of prior knowledge (terrain type, maps, beacons)
• Dealing with dynamic environments
• Sensor failures, reliability and redundancy
• Tackling difficulties in outdoor perception (sunlight, low light, smoke, rain)
• Experimental results and full system integration in real world applications
• Estimation and mapping datasets and benchmarks.