Toward a Hybrid Sensor Fusion using Probabilistic and Abstract Sensor Models
Fusion techniques are based on certain sensor models, which broadly fall into two categories: probabilistic model and abstract model. The probabilistic sensor model uses certain noise distributions on sensors (e.g., Gaussian), which is well suited for analyzing the systems’ expected performance in the average case. However, wrong assumptions on noise distributions may be in danger of being vulnerable to sensor attacks. On the other hand, the abstract sensor model uses the worst-case error bound of sensors. Thus, this model is well suited for the systems’ worst-case performance, which is highly relevant to the case of sensor attacks. In this work, we study a hybrid sensor fusion that uses both probabilistic and abstract sensor models to be able to benefit from both. We demonstrate the validation of our hybrid sensor fusion technique using an unmanned ground vehicle called Jackal.
Minsu Jo (Minsu-Jo@dgist.ac.kr)
Junkil Park (email@example.com)
Youngmi Baek (firstname.lastname@example.org )
Radoslav Ivanov (email@example.com)
James Weimer (firstname.lastname@example.org)
Sang Hyuk Son (email@example.com)
Insup Lee (firstname.lastname@example.org)
Minsu Jo, Junkil Park, Youngmi Baek, Radoslav Ivanov, James Weimer, Sang Hyuk Son, Insup Lee, “WiP Abstract: Toward a Hybrid Sensor Fusion using Probabilistic and Abstract Sensor Models,” IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), Daegu, Korea, Aug 2016.