# Martin Källman

less info
reputation
1
bio website location London, United Kingdom age 28 member for 1 year, 7 months seen Jan 28 at 18:59 profile views 7

 May15 comment Minimizing distance variance between points on a sphere Thank you, that's what I'm looking for :) May14 comment Minimizing distance variance between points on a sphere If we start at "the top" and traverse down the Y-axis in a swirl pattern, then the distance between a point and its geometrically immediate neighbours would be significantly different depending on whether or not a neighbour is the successor/predecessor on the path. I.e. I'm looking for as tight of a distribution as possible of the distances between all nodes May14 comment Minimizing distance variance between points on a sphere Perhaps the metric is ill-chosen. I'm looking to minimize entropy, maybe you have any ideas on a better one? Jul24 comment Approximating number of nodes expanded by A* search A slightly cheaper alternative could be f(h,c,ratio_n) = pow(h,c*ratio_n) where h is the heuristic function, and ratio_n is the occupied:unoccupied vertex ratio on the whole graph. Jul24 comment Approximating number of nodes expanded by A* search The cost function employed is usually the Euclidean distance, and the heuristic is usually the Manhattan distance. What about f(d,c,k,n) = pow(d,c) * k * n where d is the distance/cost function, c and k are constants determined experimentially and/or stochastically, and n is the number of occupied vertices in the rectangle that bounds the line segment start->goal? Jul24 comment Approximating number of nodes expanded by A* search Sounds interesting! I wonder if it will actually prove to be slower than the actual A* search, though? Jul24 comment Approximating number of nodes expanded by A* search I am not really too familiar with that system being a programmer... (but apparently I should be! ;) Roughly how would it be applied, and is it feasible for an online algorithm?