This repo has implementation of RRT,
RRT* and Informed RRT* algorithms for path planning.
Algorithms are stored in the algs module, and execution
lays in the main.py file. You can change which algorithm is used
and then run from the repo's root directory using :
py main.pyIn [1], the cost function is defined such that Cost(p) = Parent(p) + c(Line(Parent(p), p)).
Here, cost of a node p is merely the current distance traveled up until attained p.
The implementation of CostTreeGraph#get_cost does not perform this. Instead, the cost of a node
p is directly set in the algorithms. This way, it is not required to recursively compute the
cost of a node, thus allowing for O(1) access.
-
S.Karaman, E.Frazzoli, Sampling-based Algorithms for Optimal Motion Planning https://arxiv.org/pdf/1105.1186
-
S.M. LaValle, Rapidly-Exploring Random Trees: A New Tool for Path Planning https://msl.cs.illinois.edu/~lavalle/papers/Lav98c.pdf
-
Daichi Miyajima et al., KIKS Extended Team Description for RoboCup 2023, https://ssl.robocup.org/wp-content/uploads/2023/02/2023_ETDP_KIKS.pdf