Preliminary Design of the Dragonfly Navigation Filter

Ben Schilling,Timothy G. McGee,Ryan Mitch,Ryan Watson

2023-08-02

Abstract:Dragonfly is scheduled to begin exploring Titan by 2034 using a series of multi-kilometer surface flights. This paper outlines the preliminary design of the navigation filter for the Dragonfly Mobility subsystem. The software architecture and filter formulation for lidar, visual odometry, pressure sensors, and redundant IMUs are described in detail. Special discussion is given to developments to achieve multi-kilometer surface flights, including optimizing sequential image baselines, modeling correlating image processing errors, and an efficient approximation to the Simultaneous Localization and Mapping (SLAM) problem.

Robotics

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to design a reliable navigation filter for the Dragonfly probe, enabling it to perform multi - kilometer - level autonomous flights on the surface of Titan. Specifically, the paper mainly focuses on the following points:

1. **Challenges in Autonomous Navigation**:
- Since Titan has no external infrastructure or orbital assets, it cannot rely on GPS on Earth or other satellite systems.
- The map resolution on the surface of Titan is low and cannot provide high - precision terrain information.
- The surface environment is complex and unknown, requiring robust algorithms to handle limited prior knowledge.

2. **Application of Visual - Inertial Odometry (VIO)**:
- Using the IMU and continuous images to estimate the displacement of the aircraft solves the problem of the lack of high - resolution maps.
- By saving "breadcrumbs" (i.e., images of key positions), these positions can be revisited in subsequent flights, thereby reducing cumulative errors.

3. **Efficient Approximation of the SLAM Problem**:
- Traditional SLAM methods are difficult to implement on the selected flight processor. Therefore, the paper proposes an efficient approximation method to handle the simultaneous localization and mapping problem.
- It specifically discusses issues such as optimizing the continuous image baseline and modeling relevant image processing errors.

4. **Navigation in the Entry, Descent, and Landing (EDL) Phase**:
- The EDL phase has challenges such as initial attitude errors and high angular velocities. The paper describes how to use the same navigation filter to deal with these problems.

### Formula Summary

- **Attitude Propagation Formula**:
\[
\dot{R}_{b}^{tof} = R_{b}^{tof} [\omega_{tof}^{b|tof}]_\times
\]
where \( R_{b}^{tof} \) is the direction cosine matrix from the TOF coordinate system to the body coordinate system, and \( \omega_{tof}^{b|tof} \) is the angular velocity in the body coordinate system.

- **Position and Velocity Propagation Formulas**:
\[
\dot{r}_{tof} = v_{tof}
\]
\[
\dot{v}_{tof} = R_{tof}^{tci} a_{tci} - 2\omega \times v_{tof} - \omega \times (\omega \times r_{tof})
\]

- **Barometric Sensor Measurement Model**:
\[
P_T = P_S + P_D = P_0 \exp\left(\frac{h - h_0}{H}\right) + \frac{1}{2} \rho v_{rel}^2 C_p(\alpha, \beta)
\]
where \( P_T \) is the total pressure, \( P_S \) is the static pressure, \( P_D \) is the dynamic pressure, \( P_0 \) is the reference pressure, \( h \) and \( h_0 \) are the altitude and the reference altitude respectively, \( H \) is the scale height, \( \rho \) is the atmospheric density, \( v_{rel} \) is the relative wind speed, and \( C_p \) is the pressure coefficient.

- **Lidar Measurement Model**:
\[
y = d \bar{l}_{tof} \cdot \bar{n}_{tof} + \eta
\]
where \( d \) is the distance from the ground to the aircraft, \( \bar{l}_{tof} \) is the unit vector of the laser point in the TOF coordinate system, and \( \bar{n}_{tof} \) is the normal vector of the ground.

Through these formulas and methods, the paper aims to ensure that Dragonfly can perform precise and reliable autonomous navigation on the surface of Titan, thereby successfully completing scientific exploration tasks.