You're not trying to get absolute position from those sensors, are you? I spent far too long on that project. At least at the precision we needed, my conclusion

You're not trying to get absolute position from those sensors, are you? I spent far too long on that project. At least at the precision we needed, my conclusion was that you would want to acquire and process the images yourself - both to get a higher resolution/field-of-view, and to estimate the error associated with each position update (Kalman filtering or similar). You can also then "re-register" your position if your path travels over the same spot twice. Hopefully your application is less demanding!