Extending Navigation Service under Sensor Failures: An Approach by Integrating System Identification and Vehicle Dynamic Model

1) What problem is this paper solving?

Context: Sensor-based localization (e.g., visual positioning) fails in extreme conditions.
Core contribution: A sensor-free localization method using online System Identification (SI).
Achieved goal: Extended navigation capability without collisions during sensor outages.

2) Why is this paper important?

What changed: Autonomous systems require continuous availability even when sensors fail.
Problem created: Conventional Vehicle Dynamic Models (VDM) drift too fast to be useful backups.
Why current solutions fail: They lack real-time adaptation to the vehicle’s actual dynamics.

3) How does this paper solve it?

Contribution 1: Identifies system dynamics of powertrain and steering online.
Contribution 2: Uses identified responses to drive the VDM for positioning.
Key result: Successfully navigated a 140-meter trajectory without collision during a complete sensor failure.

🎯 Takeaway: Online system identification turns the vehicle model into a reliable backup navigator.

@inproceedings{yan2023plans,
  author = {Yan, Penggao and Hsu, Li-Ta and Wen, Weisong},
  title = {Extending Navigation Service under Sensor Failures: An Approach by Integrating System Identification and Vehicle Dynamic Model},
  booktitle = {2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)},
  year = {2023},
  pages = {630--636},
  doi = {10.1109/PLANS53410.2023.10140089}
}