The log files for laser contains range readings in each direction (1-D scan horizontal).

The first number is the time stamp of the laser which matches with the filename of the images. The time stamps of the laser scans and the images are off by about 1 second (same for all images) -- so some matching needs to be done.

1.57 and -1.57 are the angles in radians, indicating that the scan was done for 180 degrees. There are 361 readings, each separated by .0087 radians. It is followed by 361 readings in meteres, indicating the range ineach angle. (The corresponding image is a subpart of this scan).

Once you align the images (in bmp and png format) and laser scans for one image (by adjusting the 1 second timestamp difference, and cropping the laser scans lying out of image taken), it will work for images.

High Speed Obstacle Avoidance using Monocular Vision and Reinforcement Learning, Jeff Michels, Ashutosh Saxena, Andrew Y. Ng. Proceedings of the Twenty-first International Conference on Machine Learning, ICML 2005. 

http://ai.stanford.edu/~asaxena/rccar/

(Thanks to Jamie Schulte for help in collecting the laser data)