20 mins in and there are multiple typos on the board, it is hard to follow, dots, underscores, forgets to add a term in one step then adds it in the final step etc.
Really interesting ! Thank you ! Are laser guided missile using the difference of brigthness created by the laser spot on the target to help computer vision or is it time-to-flight sensor ? Could it help to light the moving target with laser spots to increase brightness gradient ?
Well the one I worked on had a 4 quadrant sensor. A portion of the reflected laser from the target was coming to the missile aperture, and the optics was purposefully designed out of focus so that the laser point on the target is mapped as a circle on the sensor. Then each quadrant of the sensor outputs voltage proportional to the amount of energy received. If the left 2 quadrants received more energy than the ones on the right, then it means the target is, lets say, on the left. Similarly, if the upper two quadrants received more energy than the lower ones, then the target is, lets say, down. According to the ratio of energies on the quadrants, you guide the missile. You try to maximize the amount of incident energy by using more powerful designators, using wavelengths that are less absorbed by the atmosphere, increasing transmittance of your optics, increasing sensitivity of your sensor. The target tries to minimize your chances by using special coatings to absorb energy, decrease surface reflection, try to make surfaces that reflect non-uniformly so that the reflected energy doesnt reach your optics but goes another way, use smoke to increase absorption at those wavelengths, create a decoy so that the missile is guided to the decoy instead etc.
There seems to be a jump at 1:02:00 where the aperture problem appears on board but is not discussed
20 mins in and there are multiple typos on the board, it is hard to follow, dots, underscores, forgets to add a term in one step then adds it in the final step etc.
How did the negative sign come at 39:50?
thank you for making this beautiful material available
Really interesting ! Thank you ! Are laser guided missile using the difference of brigthness created by the laser spot on the target to help computer vision or is it time-to-flight sensor ? Could it help to light the moving target with laser spots to increase brightness gradient ?
Well the one I worked on had a 4 quadrant sensor. A portion of the reflected laser from the target was coming to the missile aperture, and the optics was purposefully designed out of focus so that the laser point on the target is mapped as a circle on the sensor. Then each quadrant of the sensor outputs voltage proportional to the amount of energy received. If the left 2 quadrants received more energy than the ones on the right, then it means the target is, lets say, on the left. Similarly, if the upper two quadrants received more energy than the lower ones, then the target is, lets say, down. According to the ratio of energies on the quadrants, you guide the missile. You try to maximize the amount of incident energy by using more powerful designators, using wavelengths that are less absorbed by the atmosphere, increasing transmittance of your optics, increasing sensitivity of your sensor. The target tries to minimize your chances by using special coatings to absorb energy, decrease surface reflection, try to make surfaces that reflect non-uniformly so that the reflected energy doesnt reach your optics but goes another way, use smoke to increase absorption at those wavelengths, create a decoy so that the missile is guided to the decoy instead etc.
There are many mistakes in the vector notations of lecture notes compared to what is written on the board. Please correct them.
Thanks interesting optic flow applied to aereospace , lithium .Machine vision , computer vision in rocket is interesting .
1:07:39
Where’s the first part?
Here is the course playlist: th-cam.com/play/PLUl4u3cNGP63pfpS1gV5P9tDxxL_e4W8O.html
@@mitocw thanks :)