Raymond E. Arvidson

Department of Earth and Planetary Sciences

McDonnell Center for the Space Sciences

Washington University

St. Louis, Missouri 63130

Arvidson@wunder.wustl.edu

314 935 5609

fax: 314 935 4998

Paul Backes, Jeff Norris, and Eric Baumgartner

Maria Bualat, Hans Thomas, and Carol Stoker

Steve Squyres

4/19/00

1. OVERVIEW

This document is a plan for a joint field experiment in the southwestern United States involving the JPL FIDO Rover and the AMES FIDO-class K9 Rover during fiscal year 2000. Specifically, the FIDO Rover would be deployed in the field from May 6-19 to conduct tests related to use of the Athena Mars Sample Return Rover, focusing on identification of rock targets, traversing to them, conducting in-situ measurements, and collecting drill cores. FIDO will thus simulate a Sampling Rover. From May 13-16, K9 will also be on site to simulate a Scout Rover that scouts ahead and identifies and maps rock targets to determine scientific utility and the feasibility of sampling with the FIDO Mini-Corer . The intent is to simulate how two rovers would be able to work together on Mars to do an efficient job identifying, mapping, and sampling specific rocks and soils. Rover command and control will be through a single Core Operations Team (COT) located at the Jet Propulsion Laboratory. The COT will not have prior knowledge of the site, except for orbital and airborne imaging data. FIDO will be commanded using WITS, with data visualization support from the Ames Viz team, and K-9 will be commanded using the Viz software. Data visualization will be accomplished using these systems and common frames of reference to identify key features in the various data sets. A distributed group of high school students who form the LAPIS-2 Group will conduct operations for the FIDO rover from May 17-18, guiding the rover from over the horizon back to the lander and acquiring imaging data along the way. Documentation and data will be archived with the Planetary Data System. Press and public outreach will be coordinated through Mary Hardin, JPL.

2. INTRODUCTION

The exploration of Mars will certainly involve surface mobility. Currently the baseline for Mars Sample Return includes the Athena Rover and Payload for 2003 and 2005, with a focus of acquiring rock cores and soil samples and delivering the caches to an ascent vehicle for eventual return to Earth. The rover will also be able to traverse kilometers in distance after samples are delivered, conducting the exploration and discovery portions of its mission. After this first phase of surface exploration is accomplished one can imagine further surface exploration using rovers for mobility. In fact, pairs of rovers can be envisioned to explore jointly the surface and collect samples, in much the same way that pairs or groups of explorers on Earth traversed complex terrains and ensured each other’s safety.

The field trials to be accomplished from May 6-19, 2000 will involve the prototype Athena Rover, JPL’s FIDO, to simulate the complex surface operations needed to identify, get to, and sample rocks on Mars (See Table 1 for summary of activities). FIDO will be accompanied by the AMES K9 Rover for several days. During this period K9 will assume the role of the Scout Rover and FIDO will continue to simulate a Sampling Rover. The intent is to set up goals and objectives for the test and to see how well then can be met. The intent of this document is to delineate the rationale and overall set of plans. Additional documents have been prepared for the Bureau of Land Management and the relevant Department of Transportation to obtain permission to use specific sites. Permission has, in fact, been granted. Detailed plans will also be developed for command and control of the rovers and a set of test objectives. Metrics will be developed to determine the extent to which objectives have been met. Lessons learned and journal articles based on test results will be published.

The Athena Rover will be an integral part of the current baseline Mars Sample Return Mission and will have the following capabilities

• Supports the Athena science payload
• Can traverse up to 100 meters per sol
• Provides a navigational accuracy of 10% or better
• Provides high-speed computational capability and substantial data storage
• Can acquire samples and transfer them to the Mars Ascent Vehicle (MAV)

The Athena science payload will consists of the following components:

• Pancam: A high-resolution, color, stereo panoramic imager
• Mini-TES: A mid-infrared point spectrometer
• APXS: An elemental chemical analyzer
• Mossbauer Spectrometer: For analysis of minerals containing Fe
• Raman Spectrometer: For precise mineralogical identifications and point counts
• Microscopic Imager: A fine-scale (30 m m/pixel) close-up imager
• Mini-Corer: A rock coring drill and soil sampler

The Athena Rover will have several additional camera systems: Navcams atop the mast for navigation, front and rear Hazcams for hazard-avoidance maneuvering, and Bellycams under the rover body to monitor Mini-Corer operations.

FIDO is a prototype of the Athena Rover, equipped with many of the elements of the Athena Payload. It is used to simulate the complex surface operations expected of the Athena Payload, focusing on identification of rock targets, approaching the targets and conducting in-situ measurements, and drilling/verification of cores.

The instrument payload on FIDO includes a mast that is stowed on the rover deck as the vehicle is moving and deployed to 1.94 m height for acquisition of stereo imaging and spectral reflectance data. The mast head houses Pancam, a three band color (0.65, 0.75, 0.85 micrometers) imaging system capable of surveying the terrain in stereo with high spatial resolution for scientific purposes. Navcam is also included and is a low spatial resolution, monochromatic, wide field of view stereo imaging system used for traverse planning. These systems are similar to those to be flown on the Athena Rover, although the resolution of the FIDO imaging systems is not quite as good as planned for the Athena Payload and the number and wavelength coverage of the FIDO Pancam bands is less than planned for the Athena Pancam. The Infrared Point Spectrometer (IPS) on FIDO is bore-sighted with Navcam and Pancam, and acquires spectral radiance information over the wavelengths from 1.3 to 2.5 micrometers with a spectral resolution of approximately 10 cm^-1. An IPS pixel covers approximately 9 by 9 Pancam pixels. The IPS can be used both in a point mode and a mode in which a suite or raster of data are acquired to form an image cube. Mini-TES (Thermal Emission Spectrometer) will be flown on the Athena Rover, with a wavelength coverage from 5 to 20 micrometers and a spatial raster size choice of either 20 or 8 Pancam pixels. The Mini-TES was too expensive to replicate for FIDO. Thus the less expensive IPS was built to simulate joint use of imaging and spectral data.

An four degree of freedom arm is included on the front of the FIDO Rover. The end effector on the arm includes a color microscopic imager and an ⁵⁷Fe Mössbauer Spectrometer. Athena will have a arm and similar instruments, with the addition of a calibrated Alpha Proton X-Ray Spectrometer and a Raman Spectrometer. The APXS will not work in Earth’s atmosphere and the Raman Spectrometer was not ready for inclusion on FIDO for the field tests. The Mini-Corer on FIDO is an Athena prototype rock drill that is mounted separately from the arm and that can be commanded to pitch down and acquire a 0.5 cm diameter by up to 1.7 cm long cores. The equivalent dimensions for the Athena Mini-Corer will be 0.8 × 2.5 cm. Cameras mounted on the underside or "belly" of FIDO monitor drill deployment. Similar belly cameras will be used on the Athena Rover. For FIDO the core can be extracted from the rock and examined with the microscopic imager. Once a core’s presence is confirmed it can either be ejected or kept and placed in a caching tube. For the FIDO Rover, the microscopic imager and Mössbauer Spectrometer can also be placed against rock and soil targets to acquire close-up views and information on iron oxidation state and the mineralogy of iron bearing phases. For the Athena Rover all three spectrometers and the microscopic imager will be used to obtain rock and soil data. In addition, each instrument will be able to be placed against the Mini-Corer bit end to make measurements of the rock core tips. Note that the Mössbauer Spectrometer will not be used during the May 2000 tests. Only the microscopic imager will be used during arm deployments. This decision is based on radiation safety issues and the focus for the tests on sampling.

Hazard avoidance cameras on FIDO are located on the front and back of the vehicle to acquire stereo images and terrain maps of the areas to be traversed. On-board autonomous hazard-avoidance software is used to judge whether obstacles are to high to be successfully traversed over. If judged to be a hazard the software then commands the vehicle to search for and implement a traverse to go around the obstacle, while still trying to reach a waypoint designated remotely by the science team. Similar systems will be used on the Athena Rover.

For the May 2000 field trials FIDO will be commanded using WITS from JPL to simulate the main tasks expected of the Athena Rover during the sample collection phase of the Mars Sample Return Mission. These activities are summarized in the paragraphs that follow.

The primary task of the Athena Rover will be to select and collect rock and soil samples for return to Earth. Careful selection of samples is the key to obtaining the diverse sample set called for by the MSR Level 1 Requirements. The sample selection strategy relies heavily on the use of remote sensing to identify promising candidate rocks for sampling. Pancam is used to assess the morphology and color of candidate sampling targets, and Mini-TES is used to assess their mineralogy. If additional information on a prospective sample is desired, any of the arm instruments can be used: Microscopic Imager for fine-scale structure, Raman Spectrometer and Mössbauer Spectrometer for mineralogy, and APXS for elemental chemistry. During the field tests, FIDO Pancam, Navcam, and IPS data will be used to simulate this portion of Athena operations.

In addition to searching for prospective samples, all of the instruments can in principle also be used to acquire additional scientific data concerning the landing site. The scientific understanding that comes from such data has the potential to further aid sample selection. However, the priority of sampling dictates that ancillary scientific data be acquired only insofar as they do not impede the process of effective sample selection and collection.

Once a rock has been selected for sampling, the rover is positioned above it and the Mini-Corer is used to collect two or more rock cores from it. Each core is imaged with the Microscopic Imager after collection, to verify that the core has been successfully acquired and to document its appearance. At least one core from each rock sampled is viewed with the Raman Spectrometer and Mössbauer Spectrometer to thoroughly document its mineralogy and its possible organic content. During the May 2000 tests, Pancam, Navcam, Hazcam, and Bellycams will be used to help define rover positioning. The Mini-Corer will be deployed and the bit will touch the surfaces of the targets. Some drilling will be accomplished. The Microscopic Imager will be deployed onto selected surfaces and also used to verify core presence.

Figure 1 shows in flowchart form how the Athena Rover will typically be operated.

In normal operations, there are five general classes of daily operations that can take place in a sol:

• Panorama Day: Acquire remote sensing data of the scene around the rover.
• Drive Day: Move up to 100 meters in the direction of some selected target or area of interest.
• Approach Day: Move to within ~1 meter of a sampling target.
• Sample Day: Move onto a sampling target and acquire one core from it.
• Drilling Day: Acquire additional cores from a sampling target.

Each of these is discussed in detail in the following subsections.

Panorama Day: A panorama day is used when detailed remote sensing information is needed to select candidate sampling targets.

Although the focus is on remote sensing, panorama days can begin with a Microscopic Imager sequence on some "target of opportunity" that lies within the work volume of the arm. This sequence is performed quickly and without moving the rover, and is an example of an observation that provides useful additional scientific data without slowing the sample selection/collection process.

The main events of a panorama day are collection of a 360º Pancam panorama, and a Mini-TES panorama. There is considerable flexibility in how observing time and data volume can be distributed between Pancam and Mini-TES, and in how the observations can be laid out. In rare instances it may be beneficial to perform two successive panorama days from the same location, to provide a more comprehensive view of the scene.

During a panorama day, target-of-opportunity Mössbauer observations may also be made on a target within the arm’s work volume

After a panorama day, overnight target-of-opportunity observations may be made on a target within the arm’s work volume using the APXS, Raman Spectrometer, or Mössbauer Spectrometer.

Drive Day: The primary objective on a drive day is to move the rover from one location to another. In particular, drive days are used when the distance to be traversed is more than ~10 meters. Because the rover’s navigational errors can be as large as 10%, drive days typically end with a positional uncertainty that exceeds 1 meter.

The first event of a drive day can be to acquire a target-of-opportunity Microscopic Imager sequence on some target within the work volume of the arm. After the arm has been stowed, the rover drives over the commanded traverse. In order to help reconstruct the events of the traverse after it has happened, Navcam panoramas are typically acquired every 20 meters. At the end of the traverse, Hazcam, Bellycam, and a 360° Navcam panorama are acquired to document the rover’s new location. The remaining time and power for that sol are then used to acquire the best possible Pancam and Mini-TES data.

Approach Day: Approach days are used to place the rover close enough to some target rock that only one more command cycle will be required to get the Mini-Corer into position to sample that rock. Because the final move onto a rock should not be longer than about 1 meter (and because rover navigation errors can be as large as 10%), the distance traveled in an approach day is generally no more than 10 meters.

An approach day can begin with a target-of-opportunity Microscopic Imager sequence on some target within the arm’s work volume. The rover then drives to a point where the center of its coordinate system (the midpoint between the two middle wheels) is positioned ~1.5 meters away from the intended sampling target, with the rover facing toward the target. The sol concludes with Navcam, Hazcam, and Bellycam images to document the new position, and the best possible Pancam and Mini-TES data on the intended target.

Sample Day: Sample days are used to move the rover into position to take a core from a rock, and to acquire a core.

Sample days cannot begin with any target-of-opportunity observations with arm instruments, because a flight rule related to planetary protection prohibits touching a sampling site with any part of the rover other than the Mini-Corer before sampling takes place.

The first event of a sample day is to drive the rover into position to acquire a core, and to acquire Navcam, Hazcam, and Bellycam images to document the rover’s new position. The rover then downlinks these images and waits for a command. This event is called a breakpoint. During the breakpoint, images are analyzed on the ground, and one of two possible commands is sent to the rover:

• Deploy the Mini-Corer and acquire a sample. This command is issued if the images show that the rover is in the proper position for sampling.
• Execute a fallback sequence. This command is issued if the images show that the rover is not in the proper position. A fallback sequence is a pre-stored sequence, typically consisting mostly of Pancam and Mini-TES commands. It is intended to assure that the rover’s time is spent productively for the remainder of the sol while the reasons for the improper positioning are determined and corrective commands (for execution on the next sol) are generated.

In the nominal case where a command is issued to obtain a sample, the Mini-Corer is deployed and drills a core from the target rock. Once the core has been broken off, the Mini-Corer pitches upward, and the Microscopic Imager is used to image the Mini-Corer tip. These images verify successful acquisition of the core, document its appearance, and allow for assessment of bit wear. To the extent that time and power allow, the remainder of the sol can be filled out with Pancam and/or Mini-TES data. Overnight, Raman and Mössbauer data are obtained on the end of the core to thoroughly document its composition and hence allow more effective selection of subsequent samples.

Drilling Day: Because a minimum of two cores must be obtained from every rock sampled, each sample day is normally followed by one or more drilling days. The purpose of a drilling day is to acquire additional cores from a rock.

The first event of a drilling day is to cache the core obtained on the previous sol. (This is done as early in the sol as possible, to minimize the core’s exposure time and hence potential for contamination.) The next step is to acquire another image of the Mini-Corer tip to document the appearance of the Mini-Corer after caching has been completed. Next, a Microscopic Imager sequence can be obtained on a target of opportunity within the work volume of the arm.

The main events of a drilling day are acquisition of two more cores, separated by a breakpoint. The Mini-Corer is deployed, either so that it goes back down a pre-existing hole, or so that it initiates a new hole at a different position along its translate axis. A core is obtained and broken off, and the Mini-Corer is pitched up to the horizontal position. A Microscopic Imager sequence is obtained on the Mini-Corer tip, and the data are downlinked. Following this breakpoint, one of two possible commands can be issued:

• Cache the core and acquire another one. This command is issued if the first core has been obtained successfully.
• Execute a fallback sequence. This command is issued if the first core has not been obtained successfully.

In the nominal case where the first core has been obtained successfully, the core is cached immediately, and the Mini-Corer is pitched back down and translated to a new position to acquire a second core. This core is then imaged with the Microscopic Imager for verification and documentation purposes. The sol may be completed with Navcam and/or Pancam data as desired. (It may or may not be possible to acquire significant Mini-TES data at the end of a drilling day, depending on the time and power expended while drilling.) Overnight APXS, Mössbauer, and/or Raman measurements on a target of opportunity within the work volume of the arm can follow a drilling day.

The first event on any sol that follows a drilling day is to cache the second core obtained on the previous sol.

The NASA Ames Research Center K9 rover is a testbed for autonomy, navigation, instrument, and mission operations technologies developed in NASA’s various technology programs. It is kinematically identical to the FIDO. It features an electronics architecture which implements aspects planned for the 2003/2005 Athena Rover, facility instruments for high-resolution imaging and elemental analysis, wide-angle imagers for obstacle avoidance/navigation, and a modular interior and exterior layout to facilitate future integration with additional user payloads.

The instrument payload is focused on the task of performing high-resolution remote sensing of geological targets. For the May 2000 field test, this payload will consist of a set of mast-mounted imagers and an elemental analyzer collectively known as HindSight. This package consists of a stereo pair of high-resolution multi-spectral cameras (HawkEye), a co-sighted stereo pair of wide-angle monochrome imagers (WideEye), and Laser Induced Breakdown Spectrometer (LIBS). The basic layout of the HawkEye optical bench is shown in Figure 2 (the mounting of the LIBS optic is TBD). The right-hand cameras and LIBS and left-hand cameras are boresighted with each other. The azimuth axis has a full 360° range motion, and the elevation axis has -85° down to +45° up range. The elevation axis is positioned 1.5 m above the ground plane. In addition, front- and rear-facing stereo pairs are mounted 40 cm above the ground, providing stereo images of the front and rear vehicle hemispheres. These imagers are identical to the WideEye imagers

The HawkEye consists of a stereo pair of high resolution multi-spectral cameras spaced on a 27.9 cm baseline. The individual cameras utilize a 960x800 CMOS detector with 10 bits/pixel resolution and square pixel format. The cameras incorporate a 4 element filter wheel with red, green, blue, and clear filters, as well as a mechanical shutter which can be closed to provide dark current images. The optics consist of a low-distortion 35mm focal length lenses with a TBD F stop, providing a TBD depth of field with .31 mRad/pixel angular resolution. The cameras utilize an all-digital SCSI-II interface for control and image data readout. With file saving overhead included, the cameras can currently acquire a monochrome image in 5.8 sec, and a three-filter RGB image in 15.6 sec. The camera specifications are summarized in the Table 2. The HawkEye stereo pair is based on the CMOS Pro camera, manufactured by SoundVision Inc. Additional information on the camera can be found at http://www.cmospro.com. Prior to the May test, the cameras will be photogrametrically calibrated, as well as calibrated for flat field distortions and chromatic aberrations.

Table 2 HawkEye Camera Specs

Resolution	800x960 pixels	Sensitivity	10 mv/electron
Pixel Size	10.8x10.8 microns	Dark Signal	15 mv/sec @ 25° C
Fill Factor	26%	Spectral Range	1200 to 400 nm
Active Area	10.8mmx8.64mm	Exposure Control	1/1000 second increments
Readout Rate	5 MHz	Max Exposure	8 second
Dynamic Range	10 bpp	Max Pixel Rate	5 megapixels/sec
SNR	62db	Memory	1 MB DRAM
Angular Res.	.31 mRad/pixel	Power Reqmts	6 w
Stereo Baseline	27.9 cm	Vergence	0°

The WideEye stereo pair consists of a stereo pair of CMOS cameras mounted on a TBD baseline. The individual cameras consist of analog (RS170) output CMOS cameras with a 510x492 pixel resolution. The cameras are programmable via an I2C interface, allowing gain and exposure to be controlled from K9’s computer. The cameras are field-locked together, providing a synchronized video stream to the rover’s video digitizer subsystem. The digitizer can simultaneously digitize the stereo video stream with 8 bpp dynamic range and 512x480 resolution. The optics consist of a 12.0mm focal length lens with a fixed aperture of F2.0, providing a TBD depth of field. The WideEye camera specifications are summarized in Table 3. Further information on the cameras can be found at http://www.ovt.com, under the OV7410 product. The left and right WideEye cameras are boresighted with the respective left and right HawkEye cameras.

Table 3-WideEye specifications

The Laser Induced Breakdown Spectrometer (LIBS) is a remote sensing capability for measuring the elemental composition of rocks and soils from up to 10 meters away. The instrument is being developed for the Mars Instrument Development Program (MIDP) by Los Alamos National Laboratory, with David Cremers as the PI. The LIBS instrument operates by illuminating a target site with a high intensity laser, converting a small amount of site material to a plasma which radiates in the visible region of the spectrum. A spectrometer mounted on the rover is used to capture the spectra of this plasma, and the elemental composition can be inferred. The LIBS consists of a set of electronics mounted inside the rover chassis, as well as a set of optics co-mounted with the WideEye and HawkEye cameras. The rover-mounted electronics consist of a power supply for the laser, a spectrometer, and associated control electronics. The mast-mounted equipment consists of a high-power laser, a spectrometer foreoptic, and a small rangefinder. The mast-mounted equipment is connected to the rover-mounted equipment by a fiber optic and high voltage power cable.

In addition to the K9 rover, Ames will provide 3D terrain reconstruction and user interface capabilities for the COT. Stereo panoramas from both FIDO and K9 will be processed into high resolution 3D models by the ARC Stereo pipeline. These 3D models will be available on two SGI Viz workstations integrated into the COT. A science-oriented user interface, based on products developed for the Mars Polar Lander mission, will be provided. This interface will provide rapid measurement and analysis, and will be integrated with K-9 rover commanding software.

During the first part of the tests when FIDO is alone, the focus will be on testing operations associated with finding, getting onto, drilling into, and verifying core presence, as defined in Section 3 of this mission plan. For the several days when FIDO and K9 are operating jointly, FIDO will be the sampling rover, conducting imaging, IR spectroscopy, drilling, and microscopic imaging. As a baseline, K9 will be a scout rover, focusing on use of imaging and elemental composition determinations as additional information for the COT to better characterize the science site. On a best-effort basis, with possible additional control capabilities, K9 will also be used to help determine scientific interest in and the ability of FIDO successfully into targets. Based on numerous experiments with FIDO operations, a key area in which a scout rover would help is in detailed mapping of drill targets so that FIDO can approach from the optimum direction to move over and drill into a target with the fewest fine-scale maneuvers.

Baseline:

Current K9 control capabilities allow only coarse commanding of rotation and translation, with no closed-loop control based on dead reckoning. . Driving in this mode would make it very difficult to command the robot to perform the complicated motions required of K9 to provide useful additional data that could help in FIDO drilling operations. In addition, given the time required for FIDO to position and drill cores, the overall FIDO traverses will be relatively short. Therefore, as a baseline, K9 will perform the role of a scout that gathers data about the science site further afield of the lander than the sampling rover (FIDO).

At the beginning of the joint tests the two rovers will start next to one another. FIDO and K9 will jointly acquire a 360° panorama of the vicinity. The COT will define the first drilling target based upon this panorama and choose an initial direction in which to send K9. This phase should last approximately two hours for panorama acquisition and transmission, and another hour for COT analysis, decision, and sequence generation.

FIDO will be commanded to traverse to within 1.5 m of and facing the target and will continue with sampling operations as described in Section 3. Meanwhile, K9 will traverse about the science site as directed by the COT to obtain addition high-resolution imagery and elemental composition data using LIBS. K9 will focus on acquiring science data to allow the COT to define additional drill targets.

Best-Effort Basis:

With additional control capabilities allowing K9 to be commanded to drive specified distances and to turn to specified angles, a more ambitious operational scenario becomes available. K9 can now assist FIDO coring operations as follows:

At the beginning of the joint tests the two rovers will start next to one another (1-2 m apart) as if they were landed as a single package. Each rover will acquire ~180° panorama of the vicinity. The COT will then define the first target to drill based upon these combined panoramas. This phase should last approximately an hour for panorama acquisition and transmission, and another hour for COT analysis, decision, and sequence generation.K9 will then be commanded to move to Target 1 and acquire and transmit high resolution imaging data of the immediate area, in addition to acquiring elemental information with LIBS. This information would be used by the COT to decide whether or not to move FIDO to Target 1, based on whether or not the rock can be drilled by FIDO and the scientific importance of the target. The K9 traverse, data acquisition, transmission, and analysis would take approximately two hours.

If the decision is made to proceed with Target 1, K9 will be commanded to circle Target 1 and obtain stereo images from two to three additional vantage points. Terrain models generated from these data will enable the COT to determine the optimal approach for FIDO to move over the target and deploy the Mini-Corer. This latter phase should last approximately two hours, including image acquisition and transmission and analysis by the COT.

While K9 is approaching and mapping Target 1, FIDO will remain at the landing site, acquiring IPS IR spectra of regions judged by the COT (based on panorama data analysis) to be of scientific interest. Once K9 data confirm the importance and approachability of Target 1, FIDO will be commanded to traverse to within 1.5 m of and facing the target. Traverse planning will be based on the original FIDO panorama. At the end of the traverse FIDO will acquire and transmit the standard sequence of Navcam, Hazcam, and Bellycam data covering the target. Traversing , acquisition of data, and analysis by the COT will take two hours. Commands are then given to get FIDO over the target and to transmit Navcam, Hazcam, and Bellycam data. This fine-tuning to get over the target and to transmit data will take an hour. Quick-look analyses of the data by the COT will be used to decide whether or not to deploy the Mini-Corer and to drill. If it is decided to drill, the Mini-Corer force sensor will be used to determine if the bit has reached the surface, and a Bellycam will be acquired for confirmation. This will end the day for FIDO.

During the period when FIDO approaches Target 1 and deploys the Mini-Corer, K9 will investigate other potential targets selected from original panoramas. In so doing, K9 will provides additional imaging data and elemental data during its traverses, hence providing a more complete characterization of the area surrounding the landing site. These additional imaging and compositional data will be used by COT to define Target 2. Once Target 2 is selected, K9 will circle and map the target to help determine the optimal approach vector for FIDO. The sequence then continues with FIDO approaching the target and deploying the drill while K9 moves on to an additional target. This sequence is shown schematically in Figure 3. It is expected that one well characterized target will be mapped by K9 and drilled by FIDO each day for a total of three rocks.

Figure 4-Joint use of FIDO and K9.

Roles and responsibilities for personnel associated with the rover field trials are given in Table 4.

WITS

WITS, the Web Interface for Telescience, is the operations system that will be used to command the FIDO rover throughout the field test. WITS enables its users to view a variety of data products acquired from the rover, build command sequences, simulate these sequences within WITS to confirm their validity, and uplink these sequences to the rover for execution. In addition, WITS is designed to support distributed collaboration between multiple users over the Internet. This capability will be used extensively by the LAPIS 2 student participants during the field test. Words considered to be "WITS jargon" are shown in italics. WITS is also used the command the Rocky 7 rover and was used in the Mars Polar Lander mission to generate command sequences for the Robotic Arm and Robotic Arm Camera. Users of WITS typically follow a five step process when commanding FIDO, described below.

Step 1: View Downlink Data

WITS allows its users to visualize the data received from the rover in numerous ways. Images taken from all of the rover's cameras can be viewed individually in Wedge Views or in automatically generated Panorama Views if they were taken as part of a panorama by the rover's mast cameras. A user can retrieve the 3D position of any pixel in any of these images by clicking on the image in the desired location. These locations can be named and stored as targets for use in command sequences. Overhead Views of every picture taken by the rover's cameras are automatically generated, colored by elevation and by texture. All images acquired by the rover can also be viewed in a 3D View which can be viewed from any angle, or as stereo anaglyphs. In the Panorama Views, Overhead Views, and 3D Views, the current position of the rover is drawn as a 2D or 3D model as appropriate, even if the data being viewed was acquired from a previous location of the rover. WITS also allows its users to view plots of data from the Infrared Point Spectrometer and state files that describe the status of the rover.

Step 2: Build a Sequence

WITS users use the Sequence Tab of the WITS Main Window to build a sequence to be uplinked to the rover. The low level commands that are used to drive the rover are encapsulated in WITS Macros. Macro Windows allow the user to specify arguments to WITS Macro and generate a series of complicated commands automatically. Sequences can be stored to the server and retrieved by other WITS users over the Internet, enabling WITS users to collaboratively build sequences.

Step 3: Simulate the Sequence

During the sequence generation process, a WITS user can confirm the parameters provided to a macro by simulating the macro within WITS. All commands that instruct the rover to take a picture with the mast camera cause WITS to draw image footprints in the Panorama and Overhead Views. These footprints show the area that would be covered with the current parameters to the panorama macro. IPS commands draw target dots in the Wedge and Panorama Views indicating the location that will be observed by the instrument. Finally, a WITS user can visualize the predicted location and orientation of the rover at any step of the sequence in all of the open views by clicking on that step in the Sequence Window. These simulation capabilities allow WITS users to verify the sequence before sending it to the rover.

Step 4: Uplink the Sequence

When all of the WITS users collaborating on a sequence have reviewed and approved a sequence, the WITS operator at JPL will use the WITS Execution Tab of the WITS Main Window to transmit the sequence to the rover.

Step 5: Generate Sequence Reports

Once the sequence has been transmitted to the rover, a WITS operator at JPL instructs WITS to automatically generate a detailed HTML sequence report that can be accessed using a web browser. The sequence reports include:

1. A detailed listing of the macros and steps sent to the rover.
2. The complete state of the rover at the beginning of the sequence.
3. Images of WITS views used in the making of the sequence, captured at steps of the sequence to show the effects of the step.
4. Movies built from the images captured in #3 that show the rover animating through the steps of the sequence.

Participants in the FIDO field test that would like to have some "hands-on" experience with WITS before the field test should download the public version of WITS for Mars Polar Lander, which will run on any Windows-based PC. It can be downloaded from the WITS website at http://wits.jpl.nasa.gov/.

The Viz system integrates together 3D terrain reconstruction, rover simulation, and rover commanding into a single interface. A rover operator can utilize 3D terrain models derived from rover-based images to interactively plan traverses with a real-time rover simulator. Designated waypoints and visual targets can be integrated with imaging and spectral observations into a single continuous sequence for the rover to execute. During the May tests, the interface will be configured to allow more interactive commanding of the rover than would be typical during an actual Mars mission. However, functionality will still be provided for allowing motion, imaging, and spectral activities to be commanded in a sequence-driven manner.

As a baseline for the May field trial, K9 will not have an onboard executive allowing sequence execution. However, sequence execution can be simulated using the current K9 control interface, the Virtual Dashboard. The Virtual Dashboard is a graphic user interface that allows an operator to interact with a robot by sending single commands and observing telemetry. For the May field trial, a command sequence will be generated by the COT using Viz. Instead of sending the sequence directly to the robot, the K9 operator will manually step through the sequence at the Virtual Dashboard without using intermediate telemetry or imagery. The Virtual Dashboard saves a log of commands sent, thus creating a sequence execution record.

On a best-effort basis, a command executive will be implemented onboard the K9 rover, allowing sequence-driven operations.

8. SYSTEMS TO BE USED IN THE FIELD

FIDO support equipment consists of a field trailer containing power supplies, a Sun workstation, a laptop command/control computer, an Ethernet hub, the satellite modem, and miscellaneous electronic and mechanical test equipment and tools. Wireless communication between the FIDO rover and the field trailer will be accomplished using WaveLynx wireless ethernet units. The field trailer also includes a differential GPS unit that communicates with FIDO for the determination of the differential latitude and longitude of the rover relative to the base station antenna mounted to the field trailer. All FIDO rover uplink commands issued from JPL by the COT will pass through the laptop command/control computer to FIDO. All telemetry from FIDO will be stored on the command/control computer and automatically shipped to the downlink receiver located at JPL. Support vehicles at the field site consist of a 34-foot RV, two 4-wheel drive vehicles (e.g. Ford Expedition or the like), and a U-Haul trailer which will be towed by either the RV or one of the 4-wheel drive vehicles. Two gasoline generators are utilized for field trailer power in addition to the generator located within the field trailer itself.

K9 support equipment consists of an off-board power supply for bench operation and battery charging, and up to two laptop computers, one for operations and the other for software development/activity monitoring. In addition, the LIBS instrument will be commanded and sequenced from an off-board computer via a hardline. LIBS operations must be coordinated with all personnel in the field who might be exposed to emissions from the LIBS laser (Los Alamos National Laboratory will provide a laser safety plan as an appendix). The coordination will be accomplished either through a network chat session, or via a telephone connection operating over the satellite link.The Ames support equipment and operations will be housed in an Ames Hi-Cube truck and an 30ft Ames Class C RV and Ames will provide their own 120 V AC power for these systems (gasoline generator). Ames will connect to the field local area network (LAN) via a single 10BT cable and a analog telephone line if available, with a downstream hub provided in the Ames truck for K9 functions. K9 utilizes a BreezeCom 2.4 GHz Industry/Science/Medical (ISM) spread-spectrum wireless ethernet system. Prior testing during JPL ORT’s will verify RF compatibility with JPL wireless systems.

Press and publication interactions will be coordinated through Mary Hardin, JPL.

Table 1-Summary of Field Tests by Date and Activity