CS 7638 MIDTERM 1 (FALL 2025) -
PROBABILITY, HISTOGRAM FILTERS,
KALMAN FILTERS, PARTICLE FILTERS,
BICYCLE MOTION GRADED A+
1 of 54
Definition
The location probabilities in a robot's grid cell become uniform if
you move the robot infinite times with inaccurate robot motion.
Give this one a try later!
have no effect on distribution increase distribution
limit distribution localization
, Don't know?
2 of 54
Term
policy map
Give this one a try later!
Maps each cell to the
best action Maps each cell to the worst action
Maps each cell to a neutral action Maps each cell to a random action
Don't know?
3 of 54
Term
Q: Assuming a cyclical world and exact motion, what is the
probability that the robot is in cell 2 be after TWO
movements to the right?
cell [probability]: 1 [0.3] | 2 [0.4] | 3 [0.2]
Give this one a try later!
, 0.3 0.2
0.4 0.1
Don't know?
4 of 54
Definition
never overestimates the cost to reach the goal
Give this one a try later!
Admissable heuristic Particle filter
Resampling Limit distribution
Don't know?
5 of 54
Term
Q: What is the formula for the new mean after the
prediction / motion update?
Give this one a try later!
, sigma^2 + r^2, where r is the
stdev of the measurement and mu + U, where U is the
sigma is the stdev of the motion with the same units
estimated robot position as mu (e.g. 10 m).
Z = H * x, where H is n x D,
where n is the number of x' = F * x, where F is DxD
position elements, and D is the
number of dimensions
Don't know?
6 of 54
Definition
The probability of a robot's location shifts with its
movement. This assumes there is 0 noise.
Give this one a try later!
a* search resampling
localization exact motion
Don't know?
7 of 54
PROBABILITY, HISTOGRAM FILTERS,
KALMAN FILTERS, PARTICLE FILTERS,
BICYCLE MOTION GRADED A+
1 of 54
Definition
The location probabilities in a robot's grid cell become uniform if
you move the robot infinite times with inaccurate robot motion.
Give this one a try later!
have no effect on distribution increase distribution
limit distribution localization
, Don't know?
2 of 54
Term
policy map
Give this one a try later!
Maps each cell to the
best action Maps each cell to the worst action
Maps each cell to a neutral action Maps each cell to a random action
Don't know?
3 of 54
Term
Q: Assuming a cyclical world and exact motion, what is the
probability that the robot is in cell 2 be after TWO
movements to the right?
cell [probability]: 1 [0.3] | 2 [0.4] | 3 [0.2]
Give this one a try later!
, 0.3 0.2
0.4 0.1
Don't know?
4 of 54
Definition
never overestimates the cost to reach the goal
Give this one a try later!
Admissable heuristic Particle filter
Resampling Limit distribution
Don't know?
5 of 54
Term
Q: What is the formula for the new mean after the
prediction / motion update?
Give this one a try later!
, sigma^2 + r^2, where r is the
stdev of the measurement and mu + U, where U is the
sigma is the stdev of the motion with the same units
estimated robot position as mu (e.g. 10 m).
Z = H * x, where H is n x D,
where n is the number of x' = F * x, where F is DxD
position elements, and D is the
number of dimensions
Don't know?
6 of 54
Definition
The probability of a robot's location shifts with its
movement. This assumes there is 0 noise.
Give this one a try later!
a* search resampling
localization exact motion
Don't know?
7 of 54
