
Concept and project objectives
The Flash Optical Sensor for Terrain Relative Robotic Navigation (FOSTERNAV) project is a contribution to the strengthening of the European position for space exploration.
It targets the maturation of knowledge and technologies for relative navigation sensors. Several space exploration missions are foreseen at an international level that will
bring back Human Beings to the Moon, bring them to Mars and allow bringing back on Earth samples collected on celestial bodies. To meet future missions’ terms of
requirement, new Guidance Navigation and Control (GNC) technologies are needed. The guiding objectives of the FOSTERNAV project are threefold:
- development of a novel optical sensor architecture merging imaging technologies and time-of-flight measurements,
- applicability evaluation of the sensor for object relative robotic navigation according to space mission scenarios,
- design and realization of standardized assessment protocols reproducing reliably and realistically missions' conditions.
The sensor concept is to merge the latest research notions of laser rangefinder, imaging Light Detection and Ranging device (LIDAR) for
space application, and elements having demonstrated technological supremacy in other fields such as optical telecommunication, security
sensors or robotic. Such innovative solution for GNC systems, further referred as "flash optical sensor", offer improved accurate and
robust control of position and attitude of spacecrafts. Its applicability to – for example – soft and precise landing of
sensitive scientific payload and humans on extraterrestrial bodies (Figure 1) will be illustrated by the assessment protocols fulfilled
in the project.

Figure 1. Typical application of the flash optical sensor: soft-landing.
Main leading ideas
Current technologies for GNC sensors do not provide the precision or the integrated functionalities that will be requested for future robotic, cargo or manned exploration missions,
involving precise soft-landing and in-orbit rendezvous. Mapping missions (e.g. Selene, Chandrayaan, Lunar Reconnaissance Orbiter) are not anticipated to be capable of resolving
surface features below the meter range necessary to definitively identify surface hazards. Future exploration mission phases will rely on precise object relative navigation sensors
able to supply measurement at the sub-meter resolution. Entry, descent and landing (EDL) mission phases impose demanding requirements. Present landing accuracy of automatic spacecraft
is about 10 km on Mars and 1 km on the Moon. It is performed without means for correction with reference to local obstacles. New concepts for on-board robotic GNC sensors are required
to meet challenging future missions' terms of reference. Technologies considered for EDL or rendezvous GNC sensors are key enablers to allow autonomous navigation, and safe and precise
landing on celestial bodies or docking in any illumination conditions and harsh environment.
The most efficient way to fulfil missions' requirements is by using sensors providing as much as possible self-contained 3D object-relative and highly precise information, including
ground-relative attitude, altitude, range, velocity, surface images, etc. The goal of FOSTERNAV is the development and the applicability evaluation of sensor architecture for EDL and
rendezvous mission phases. The flash optical sensor will provide to on-board landing spacecraft GNC systems or to in-orbit docking control systems, real time 3D observable measurements
and highly valuable navigation information. The goal of self-contained sensors is to supply expected information for relative navigation ideally without the support of another on-board,
on-orbit or on-earth instrument or additional processing resource.
In Europe, optical GNC sensors based on stereo cameras equipped with laser illuminators and scanning LIDAR technologies have been extensively studied for proximity operations, such as,
LIDAR-based GNC for Rendezvous and Landing LiGNC, LIDAR breadboard development ILT, and Flight experimentation on PLGTF [1]. A third new technology is emerging for optical GNC sensors,
more intensively in the U.S.A. for now. This technology is often denominated in the literature as: “flash LIDAR”. NASA has placed high expectations on flash LIDARs as an enabling robotic
navigation technology [2]. The related studies and experimentations take a large part of the development efforts and tests of the sub-programme considering sensor architectures for proximity
operations in the ALHAT (Autonomous Precision Landing and Hazard Detection Avoidance Technology) NASA programme. In Europe, flash LIDARs for space applications have received so far
comparatively lesser attention. In space vocabulary terms, this corresponds to low Technology Readiness Level (TRL) of 1 to 2 for flash LIDAR in Europe [3]. To catch up in the development
and the operation of such type of sensors in Europe, FOSTERNAV aims at developing a sensor providing highly valuable guidance, navigation and control information based on the flash LIDAR
concept. The project aims at constructing a sensor platform with expected TRL of 4.
While sensors grouped under the family term “flash LIDAR” have architectures roughly comparable to the sensor's architecture investigated in the project, it has distinctive characteristics
at the same time. The sensor features that will be introduced in FOSTERNAV are distinctly different from sensor usually placed in the category of flash LIDAR. The term “flash” is used in
opposition to the term "scanning". In present flash LIDARs, the image of a surface is captured in one shot. In a scanning LIDAR, the overall surface is scanned by a laser beam in several
laser shots (Figure) emitted by the laser source and by a complex scanning mechanism.

Figure 2. Scanning versus flash LIDAR.
The solution envisaged in FOSTERNAV has the following distinctive characteristics (Figure 3):
-
complex optomechanical scanning sub-system is absent, allowing to reduce drastically 3D image acquisition time,
-
matrix detector (or focal plane array) is used and coupled with the laser source modulated continuously by a pseudo-random sequence.
The absence of the scanner mechanism in "flash LIDAR" decreases significantly the overall mechanical and operational complexity in comparison with scanning LIDARs. To avoid the scanning mechanism
complication, other typologies use several illumination sources at the same time to scrutinize a given surface of an object and provide attitude or 3D object-relative navigation information. Moreover,
the sensor architecture investigated in the project makes use of building-blocks and presents characteristics that are not yet found in flash LIDARs.

Figure 3. Main characteristics of the flash optical sensor architecture.
Scientific and technical objectives
The scientific and technical objectives of FOSTERNAV aim at progressing beyond the state-of-art of sensor architectures for proximity operations and increasing the competitiveness of Europe,
particularly on remote sensing expertise, applicability assessment know-how and technologies related to object robotic relative navigation. FOSTERNAV's objectives are:
-
To provide specifications for flash optical sensor and define applicability assessment protocols extracted from the investigation of real exploration future missions term of requirement.
- To design and assemble a prototype of the flash optical sensor obeying the specifications of point 1 and ready to be used in applicability test facilities.
- To assess the operation and capabilities for relative navigation of the flash optical sensor in ground-test facilities (Figures 4 and 5) reproducing down-scaled dynamics and exploration conditions
of missions with either soft-landing or docking phases.
- To disseminate information including evaluation results on the sensor concept, the project approach in the interested communities and ensure the exploitation in terms of availability of the
technology for implementation by European industry and research.
The main ideas guiding the work are generated by the future requirements for space vehicles that will be designed for future planetary explorative missions to Mars (e.g. Mars Sample Return [4]),
to the Moon or to asteroids (e.g. Marco Polo [5]). These missions are a key component of the near-term European space exploration vision and also of future planned international space missions.
The ideas developed in this project are also related to the current state-of-the-art of sensor families considered for on-board guidance (for example, in the current on-going harmonization process
between ESA and Eurospace for AOCS sensors and actuators), navigation and control systems for future robotic, cargo or crewed exploration spacecrafts. The flash optical sensor will be capable of
supplying to spacecraft GNC systems local surface normal relative position (or range), attitude and velocity in approaching phases of celestial body or another spacecraft.

Figure 4. Schematic view of the Testbed for Robotic Optical Navigation (TRON) at DLR.

Figure 5. Free-range high speed testing at airfield near Bremen.
Work performed since the beginning of the project
For the first phase of the project the work performed can be classified in three categories:- Derivation of the specifications of the flash sensor from missions’ terms of requirement while taking into account current limitations of the technologies entering in the imaging LiDAR architecture. Several planned missions have been investigated by the project team: Mars Sample Return, Marco Polo and Lunar landing scenarios.
- Part of the preliminary design of the sensor prototype.
- Description of the test facilities and definition of test plan.
Main results achieved so far
-
Extensive review of missions involving either rendezvous or landing.
-
Following the extended review of missions’ scenari and requirements three modes of operation have been defined for the sensor. In short, they permit to optimize the use of the limited energy available for the sensor. This is especially important for the part of this energy that is converted in optical power by the illumination head. The general guideline is to maintain the optical power available in the smallest divergence angle to ensure the best possible signal-to-ratio on the sensor’s detector and consequently to maintain the statistical error of the measurement of the distance between the sensor and the target as small as possible.
-
In close relation with the three modes of operation, a sensor architecture has been studied and confirmed. It is innovative compared with other flash sensor from the point of view of the controllable target illumination chosen (modulated continuous wave multiple lasers source) and the type of matrix detector used (integrated demodulation pixels).
-
Elaboration of sensor models allowing at the same time full GN&C system simulation and the support of the design of the sensor.
The definition of the model varies whether it is of interest for experts in image processing or GN&C systems (denominated imager model), or for scientists designing LiDAR
systems (called performance model). For the
first group it is capital to get an image at the
output of the model while for the second it is
more about calculating the optical power budget
emitted by the laser head and then reflected
back and detected by the sensor receiver.
-
Determination and detailed description of the
test facilities that will be used to assess the
sensor’s prototype.
Expected final results and
their potential impact and use
The main final expected results of the project are:
- To identify and summarize the key requirements coming from missions’ planners for future missions that might involve flash sensors.
- To derive from the missions’ requirements a concept for the flash sensor that may potentially fulfill them.
- To build a functional sensor prototype based on an innovative flash LiDAR architecture that can be tested on real GN&C testbenches.
- To demonstrate that this
sensor architecture is
applications for rendezvous and
landing space applications.
- To generate a database of DEM coupled with trajectory data of the sensor for further investigations in future activities related to flash optical sensors. The aim is to constitute a database of 3D images acquired while the prototype is moved on test-benches to allow in the future the development of a complete system including software elements comprising for example algorithms for hazard recognition and detection, GN&C information extraction, craft attitude and motion compensation, etc.
Following the demonstration of applicability of the sensor architecture, the expected long technical and economic impact expected is to trigger larger development activities that would allow the flash technology to be further developed in Europe to a level comparable with the development of the most advanced flash solutions from the USA.
From the technical material analysed in the first phase of the project, five main directions for larger development activities are perceived. These future activities should cover the following scientific subjects:
- Development of more efficient and miniaturized continuous-wave laser head.
- Development of imager technology for space and for long range applications. In FOSTERNAV, an imager developed for motion detection for video games or microscopy is used. New imagers with better sensitivity, in-pixel demodulation capability to extract distance from phase difference and able to sustain space vibrational, thermal and radiation should be developed.
- Development of imager for any semi-autonomous or fully-autonomous systems such as driver assistance systems.
- Development targeting a drastic miniaturization of system’s building-blocks such as: the opto-mechanical part (transmitter and receiver).
- Development of the algorithms needed to process efficiently DEM and to allow data fusion of flash sensors and other GN&C sensors (IMU, accelerometer, passive camera, etc.).
With such activities Europe might compensate seriously the backwardness observed with the most advanced nations.
References
1. PLGTF: ESA Precision Landing GNC Test Facilities
2. http://www.nasa.gov/exploration/home/alhat-project.html
3.
http://en.wikipedia.org/wiki/Technology_readiness_level
4.
http://www.esa.int/esaMI/Aurora/SEM1PM808BE_0.html
5.
http://sci.esa.int/science-e/www/area/index.cfm?fareaid=108
|