Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AlertHandler: double-checking for legitimacy of an object close to wall edges. #18

Open
tsirif opened this issue Jul 3, 2014 · 5 comments
Assignees

Comments

@tsirif
Copy link
Member

tsirif commented Jul 3, 2014

It is common that due to tf arithmetical errors or errors at calculation, an alert is depicted twice in two different points in space because raycasting towards the alert's direction hits on the edge of a wall marginally.

@tsirif
Copy link
Member Author

tsirif commented Jul 8, 2014

Can be solved in conjunction with Sensor Coverage: Every time coverage is updated, objects in primary sensor's range will be checked. Their probability is to be punished by a standard percentage (depends on the object type as a consequence of the differences in sensitivity of detector). If an object's probability drops under a certain threshold (let's say that 25% is fine), it is deleted. This dictates also that Alert Handler will accept only alerts whose probability is above this certain threshold.

@tsirif tsirif assigned tsirif and konstantinos41 and unassigned tsirif and konstantinos41 Dec 11, 2014
@tsirif
Copy link
Member Author

tsirif commented Dec 16, 2014

Examine a solution with kinect: Alert messages should include the depth of poi's pixel. So we can check if the distance of the object and us is consistent with poi's depth.

@tsirif
Copy link
Member Author

tsirif commented Dec 16, 2014

Pois from usb cameras should be checked again with kinect (??)

@tsirif
Copy link
Member Author

tsirif commented Dec 16, 2014

Examine if we can have a kinect forwards and a kinect backwards, so we can have total coverage all around.

@czalidis
Copy link
Member

Αυτά τι τα γράφεις εδώ; Δεν έχουν καμία σχέση με τον κώδικα, γράψτα στο
trello!

On Tue, Dec 16, 2014 at 9:11 PM, Christos Tsirigotis <
[email protected]> wrote:

Examine if we can have a kinect forwards and a kinect backwards, so we can
have total coverage all around.


Reply to this email directly or view it on GitHub
#18 (comment)
.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants