[Discussion] AutoCAD Curve gemetry - Calculating an average slope

[jammie]

I'd like to take an AutoCAD curve (polyline) & calculate the aproximate overall average slope of it. Attached is my code so far which works

(defun weightedAverageSlopeOfCurve (<obj>)
  
  ;Some function variables  
  (setq chainage 0.0 interval 0.1 tempList nil orig (list 0 0 0 ) maxIterations (fix 1e5) average 0.0)

  (setq startChainage (vlax-curve-getDistAtParam <obj> (vlax-curve-getStartParam <obj>))
	endChainage  (vlax-curve-getDistAtParam <obj> (vlax-curve-getEndParam <obj>))
	)

  ;Define the length of the curve
  
  (setq curveLength (-  endChainage startChainage  )
	chainage  startChainage)

  ;Caclulate the number of iterations
  
  (setq iterations (fix (/ curveLength interval)))

;;Take some loading off the CPU if too many iterations required. Revert to maxIterations 
  (if
    (> iterations maxIterations)
    (setq interval (/ curveLength maxIterations) iterations maxIterations)

    )

  (repeat iterations

    (setq dy/dx (vlax-curve-getFirstDeriv <obj> (vlax-curve-getParamAtDist <obj> chainage))
	  slope (angle orig dy/dx))

    (setq average (+ average (/ slope iterations)))

    (setq chainage (+ chainage interval))
    
    )
  average
  )

Has anybody any thoughts on the methodology or suggestions for a different approach?

[BIGAL]

your question is to vauge what is an average slope ? post example

[jammie]

your question is to vauge what is an average slope ? post example

Hi BIGAL,

Sorry for not getting back to this sooner. I'll try and be a little clearer, average slope probably was the wrong term to use.


Taking an AutoCAD curve, in this case a polyline, I am intereseted in determining the average vector travelled going from the start point to the end point of the polyine.

My original thinking was to break the curve down into a series of segments, calculate the slope at each segment & from that calculate an overall average slope / vector travelled

Regards

Jammie

[BIGAL]

Wont the average of a arc just be the tangential slope at the mid point ?

[eldon]

Or even the line joining the start point to the end point

However the vectors vary along the way, they start at the start point and they arrive at the endpoint.

[jammie]

@BIGAL I hadn't considered that. I'll have to do a bit of testing to see how if/how I can incorporate this.

@eldon That was my thinking with vectors varing along the length of the curve. Every segment of the curve has a weighting on the overall average of the curve.

Perhaps the following could work?

1.    Determine the number of curve segements  			(n)
2.    Find the tangent @ each mid segment         		(m)
3.    Find length of that segment              			(l)
4.    Determine the magnitude of the vector (l*m) 		(s) 
5.    Add all the results & divide by total length of curve

Thanks for the help so far

Jammie