Autonomous and non-autonomous dynamic model based navigation system for unmanned vehicles

A navigation system including a vehicle dynamic model (VDM) that serves as the main process model within a navigation filter is described. When used in an unmanned aerial vehicle (UAV), the navigation system may work in communication with inertial measurement units (IMUs) and environment dependent sensors such as GNSS receivers. Particularly, the navigation system is beneficial in the case of GNSS signal reception outages, where conventional IMU coasting drifts quickly. Yet, the navigation system may also be employed in other scenarios, for example during GNSS presence for improved positioning, velocity and attitude determination, or in combination with GNSS when no IMU is available by design or due to a failure. In the navigation system, a solution to VDM equations provides an estimate of position, velocity, and attitude, which can be updated within a navigation filter based on available observations, such as IMU data or GNSS measurements.

A Micro Aerial Vehicle with Precise Position and Attitude Sensors

Romain Mabillard, Martin Rehak, Jan Skaloud

This study shows the potential of navigation technologies in the field of position and orientation determination on a micro aerial vehicle (MAV), which weight does not exceed 5 kg. Although the MAV systems feature high flexibility and capability of flying into areas that are inhospitable or inaccessible to humans, the lack of precision in positioning and attitude estimation on-board decreases the gained value of the captured imagery. This limits their mode of operation to indirect georeferencing. This paper pre-sents the development of a low cost MAV with navigation-sensor payload that shall assure a position and attitude control with accuracy from which either direct or integrated sensor orientation can benefit. After describing the hardware configuration and the synchronization of all measurements we present a case study that evaluates the performance of the positioning component and its application on integrated sensor orientation without ground control. There we show that thanks to the implementation of a multi-frequency, low power GNSS receiver, the system can potentially attain the mapping characteristics of much larger platforms flown on man-operated carriers while keeping the sensor size and weight suitable for MAV operations. The attitude accuracy of the developed board hosting several MEMS-IMUs is evaluated dynamically on a terrestrial vehicle using a reference (navigation grade) INS. Although this method offers continuous evaluation of the orientation accuracy and the obtained results are satisfactory with respect to the foreseen operations, this performance remains to be confirmed in a flight.

E Schweizerbartsche Verlagsbuchhandlung2014

Optimal Stochastic Sensor Error Modeling based on Actual Impact on Quality of GNSS-INS Integrated Navigation

Stéphane Guerrier, Mehran Khaghani, Jan Skaloud

Proper modeling of stochastic errors in inertial sensors plays a crucial role in the achievable quality of GNSS-INS integration especially with low-cost inertial sensors. Generalized Method of Wavelet Moments (GMWM) can model the underlying process for such errors with arbitrarily complex structure to obtain close match to the observed errors in terms of wavelet variance. In comparison to the widely used and IEEE adopted error modeling using Allan Variance, this method provides consistent estimation, ability to estimate parameters of composite stochastic models of much higher complexities, and considerably easier usage. However, the level of improvement in navigation quality does not necessarily grow proportionally to the fidelity of the error models. Therefore, opting for unnecessarily complex models may only increase computational load with no tangible gain in navigation quality. On the other hand, converging estimation of higher complexity models, generally speaking, requires longer estimation periods and higher dynamics to improve observability of error states. This implies yet another motivation to find an optimal sensor error model avoiding unnecessary complexities. In this paper, we employ two methods to investigate the effect of model complexity on integrated navigation performance. Firstly, a covariance propagation is performed on static conditions, as is the standard scenario in error analysis of inertial navigation systems. Afterwards, an emulation study is performed based on a real Unmanned Aerial Vehicle (UAV) flight and error signals of a Navchip V2 Inertial Measurement Unit (IMU). Results of both methods are in general agreement, and suggest that more complex models in general provide higher accuracy for the navigation system and a more consistent covariance prediction by the navigation filter. This difference is, however, more noticable only in GNSS outages of longer duration (tens of seconds). However, the benefits of more complex models may be only marginal in other applications, depending on the duration of inertial coasting and availability of other sensory data such as GNSS observations.

2019

Modeling and Processing Approaches for Integrated Inertial Navigation

Yannick Stebler

The challenge of estimating the position, velocity and orientation in space in a precise and reliable way, at any time, with and without reception of satellite signals, is the core subject of this dissertation. To this end, the use of Bayesian filters which fuse outputs from autonomous inertial navigation with satellite positioning is a well-accepted and largely proven approach. The quality of integrated systems is mainly driven by the errors affecting the inertial sensors. This research intends to improve the navigation accuracy of INS/GNSS by proposing and investigating novel approaches at two levels. First, a new estimation framework is developed that allows to model complex composite stochastic processes. We consolidate the proposed estimator on a theoretical basis and validate it through simulations and experiments. Results show the ability of our method to estimate models for which other conventional approaches (e.g. Allan variance and likelihood-based estimators) fail, thereby supporting the challenging stage of navigation filter design. Second, we investigate filter designs accounting for inertial sensor redundancy at observation and state levels. The benefits brought by such filters in terms of navigation accuracy and adaptive modeling of sensor noise are discussed in the context of experiments. For that purpose, a redundant MEMS-based inertial navigation system was designed and operated on a vehicle. Compared to classical single-IMU based filters, we found a significant bounding of the position, velocity and attitude error when operating redundant inertial systems. Contrary to single-IMU/GNSS systems, the redundant configuration is able to self-evaluate the level of system noise and thus to catch the effects of the dynamics. The improved performance and robustness is attractive for many applications requiring reliable and accurate trajectory determination.

EPFL2013

Optimal Stochastic Sensor Error Modeling based on Actual Impact on Quality of GNSS-INS Integrated Navigation

Stéphane Guerrier, Mehran Khaghani, Jan Skaloud

2019