962 
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
not included in the statistical analysis. For the rest of the 71 
pairs, covering a 1.6 km traverse, the success rate was 59 
percent after pre-screening. This result is not as good as results 
for the rocky outcrop area because the characteristics of bushes 
are sufficiently different from the rock shapes for which our 
software was designed. Despite the lower success rate, this 
software still achieved much better localization results than VO 
only. The 1.6 km traverse was evaluated as two separate parts 
because the corresponding VO data were processed separately. 
For the first part of the traverse (1.2 km), the relative 
localization accuracy was reduced from 19.7 to 4.1 percent (see 
Figure 14). The relative localization accuracy for the second 
part of the traverse (0.4 km) was refined from 9.9 to 8.7 percent. 
It is to note that in the traverse shown in Figure 14, VO failed at 
7 positions due to a lack of sufficient image overlap. The large 
localization error (19.7 percent) is predominately wheel 
odometry error. Integration of BA and VO significantly reduced 
this error. 
Figure 14. Test results for traverse of bushy area near Silver 
Lake, CA (units: m) 
5. CONCLUSIONS 
During the MER mission, the rovers are primarily localized on 
board by IMU, wheel-odometry, and sun-sensing technologies. 
VO technique has been effectively applied onboard over short 
distances to correct slippage errors. The BA method has been 
performed on Earth to achieve a high-accuracy solution of the 
entire rover traverse. Localization based on integrated BA and 
VO based has greatly supported mission operations for safe 
navigation and for achieving scientific and engineering goals. 
We have developed a new approach to autonomous localization 
for long-range rover traverses for future rover missions. This 
new approach integrates VO and BA with the expectation of 
achieving high efficiency and full automation. In particular, an 
automatic cross-site tie-point selection method has been 
developed to enable the BA to be autonomous. Test results 
using MER’s Spirit rover data as well as field test data acquired 
at Silver Lake, California, have verified the effectiveness of our 
autonomous BA software. This software could be used for 
onboard autonomous Mars rover localization in a rock-abundant 
landing site (like MER’s Spirit landing site) in future Mars 
rover missions. 
ACKNOWLEDGEMENTS 
This work was partially performed at the Jet Propulsion 
Laboratory, California Institute of Technology, under a contract 
with the National Aeronautics and Space Administration 
(NASA). Funding for this research by the NASA Mars 
Exploration Program and the NASA Mars Technology Program 
is acknowledged. 
REFERENCES 
Arvidson, R.E., R.C. Anderson, P. Bartlett, J.F. Bell, III, D. 
Blaney, P.R. Christensen, P. Chu, L. Crumpler, K. Davis, B.L. 
Ehlmann, R. Fergason, M.P. Golombek, S. Gorevan, J.A. Grant, 
R. Greeley, E.A. Guinness, A.F.C. Haldemann, K. Herkenhoff, 
J. Johnson, G. Landis, R. Li, R. Lindemann, H. McSween, D.W. 
Ming, T.Myrick, L. Richter, F.P. Seelos, IV, S.W. Squyres, R.J. 
Sullivan, A. Wang, and J. Wilson, 2004. Localization and 
physical properties experiments conducted by Spirit at Gusev 
Crater. Science, Special Issue on MER 2003 Mission, 305(5685), 
pp.821-824, doi: 10.1126/science. 1099922. 
Di, K., F. Xu, J. Wang, S. Agarwal, E. Brodyagina, R. Li, L. 
Matthies, 2008. Photogrammetric Processing of Rover Imagery 
of the 2003 Mars Exploration Rover Mission. ISPRS Journal of 
Photogrammetry and Remote Sensing, 63, pp. 181-201, 
doi: 10.1016/j.isprsjprs.2007.07.007. 
Li, R., K. Di, A.B. Howard, L.H. Matthies, J. Wang, and S. 
Agarwal, 2007b. Rock Modeling and Matching for 
Autonomous Long-Range Mars Rover Localization. Journal of 
Field Robotics, 24(3), pp. 187-203. 
Li, R., K. Di, J. Wang, X. Niu, S. Agarwal, E. Brodayagina, E. 
Oberg, and J.W. Hwangbo, 2007a. A WebGIS for Spatial Data 
Processing, Analysis, and Distribution for the MER 2003 
Mission. Photogrammetric Engineering and Remote Sensing, 
Special Issue on Web and Wireless GIS, 73(6), pp. 671-680. 
Li, R., S. W. Squyres, R. E. Arvidson, B. A. Archinal, J. Bell, Y. 
Cheng, L. Crumpler, D. J. Des Marais, K. Di, T. A. Ely, M. 
Golombek, E. Graat, J. Grant, J. Guinn, A. Johnson, R. Greeley, 
R. L. Kirk, M. Maimone, L. H. Matthies, M. Malin, T. Parker, 
M. Sims, L. A. Soderblom, S. Thompson, J. Wang, P. Whelley, 
and F. Xu, 2005. Initial Results of Rover Localization and 
Topographic Mapping for the 2003 Mars Exploration Rover 
Mission. Photogrammetric Engineering and Remote Sensing, 
Special issue on Mapping Mars, 71(10), pp. 1129-1142. 
Maimone, M., Y. Cheng, and L. Matthies, 2007. Two years of 
Visual Odometry on the Mars Exploration Rovers. Journal of 
Field Robotics, 24(3), pp. 169-186. 
Matthies, L., M. Turmon, A. Howard, A. Angelova, B. Tang, 
2005. Journal of Machine Learning Research. 1, pp.1-48.