+ All Categories
Home > Documents > FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: –...

FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: –...

Date post: 14-Dec-2015
Category:
Upload: raegan-brimm
View: 215 times
Download: 0 times
Share this document with a friend
Popular Tags:
18
FastSLAM Algorithm As interpreted by Aaron Lee
Transcript
Page 1: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

FastSLAM Algorithm

As interpreted by Aaron Lee

Page 2: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Figure 13.2

• For each particle:– “Retrieval”/”Prediction”• Move particle

– “Measurement update”• For each feature seen by the particle, update its mean

and covariance– “Importance weight”• Calculate how likely this particle is

• Resample• Duplicate and delete particles based on goodness

Page 3: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.
Page 4: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Page 5: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Page 6: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Page 7: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Set default importance if new feature

Page 8: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Set default importance if new feature

Pick the feature that corresponds bestto the measurement

Page 9: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Set default importance if new feature

Pick the feature that corresponds bestto the measurement If is a new feature, increase number by 1

Page 10: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Set default importance if new feature

Pick the feature that corresponds bestto the measurement If is a new feature, increase number by 1

If new feature, set mean to what measurement said.Remember that we have seen this feature once

Page 11: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

Get next particle

Move particle

Loop through all features in theparticle and determine how welleach of them correlate with themeasurement.

Set default importance if new feature

Pick the feature that corresponds bestto the measurement If is a new feature, increase number by 1

If new feature, set mean to what measurement said.Remember that we have seen this feature once

If we have seen this before, update mean and covariance.

We have seen this feature once more.

Page 12: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

Page 13: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

However, if we should have seen it, rememberthat we should have and didn’t

Page 14: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

However, if we should have seen it, rememberthat we should have and didn’t

If we see a feature less than 50% of the timethen throw it away

Page 15: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

However, if we should have seen it, rememberthat we should have and didn’t

If we see a feature less than 50% of the timethen throw it away

We are done with this particle, so move it totemporary storage place

Page 16: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

However, if we should have seen it, rememberthat we should have and didn’t

If we see a feature less than 50% of the timethen throw it away

We are done with this particle, so move it totemporary storage place

Empty out main particle storage place

Page 17: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

If we didn’t see the feature, don’t changeits properties

However, if we should have seen it, rememberthat we should have and didn’t

If we see a feature less than 50% of the timethen throw it away

We are done with this particle, so move it totemporary storage place

Empty out main particle storage place

Now we resample. This involves selecting M particles to copy from the temporary storage place to the main storage. The probability of any given particle of getting selected is based on its importance factor. w=target/proposal

Page 18: FastSLAM Algorithm As interpreted by Aaron Lee. Figure 13.2 For each particle: – “Retrieval”/”Prediction” Move particle – “Measurement update” For each.

• End.


Recommended