Scribd
Upload
7 views37 pages

51 Map Old

The document discusses various algorithms for path planning in robotics, including probabilistic roadmaps, visibility graphs, and search algorithms like Dijkstra and A*. It emphasizes the importance of collision-free motion planning in environments with obstacles and outlines methods for both offline and online planning. Additionally, it covers different search techniques such as BFS, DFS, and A* for efficient pathfinding.

Uploaded by

f20213000
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views37 pages

51 Map Old

The document discusses various algorithms for path planning in robotics, including probabilistic roadmaps, visibility graphs, and search algorithms like Dijkstra and A*. It emphasizes the importance of collision-free motion planning in environments with obstacles and outlines methods for both offline and online planning. Additionally, it covers different search techniques such as BFS, DFS, and A* for efficient pathfinding.

Uploaded by

f20213000
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 37

AI for Robotics

Path Planning
DR. ABHISHEK SARKAR
MECHANICAL ENGG.
BITS-PILANI, HYDERABAD
Path
• Probilistic Roadmap
⚫ Introduction

⚫ Dijkstra

⚫ A*

⚫ PRM

2
Path Planning

• 3D path planning algorithms include


– visibility graph which works by
connecting visible vertexes of
polyhedron,
– random-exploring algorithms such
as rapidly exploring random tree,
Probabilistic Road Map,
– optimal search algorithms (such as
Dijkstra’s algorithm, A∗, and D∗).
3
Environment perception
and Modeling

• Robots are expected to perform tasks in workspaces populated by


obstacles
• Autonomy requires that the robot is able to plan a collision-free
motion from an initial to a final posture on the basis of geometric
information
• Information about the workspace geometry can be
– entirely known in advance (off-line planning)
– gradually discovered by the robot (on-line planning)

5
Visibility Graph
• A visibility graph is a graph
whose vertices include the
start, target and the
vertices of polygonal
obstacles.

6
Visibility Graph
• Its edges are the edges of
the obstacles and edges
joining all pairs of vertices
that can see each other.

7 https://www.cs.columbia.edu/~allen/F19/NOTES/lozanogrown.pdf
Grown Obstacles
• The Minkowski sum of two
convex polygons P and Q of
m and n vertices
respectively is a convex
polygon P + Q of m + n
vertices.
• P⊕Q = { p+q | p ∈P, q∈Q }

8
Path planning algorithm
• Then the cell decomposition can be stated as follows:
– Divide F into connected regions called cells.
– Determine which cells are adjacent and construct an adjacency
graph.
• The vertices of this graph are cells, and edges join cells that have
a common boundary.

10
Motivation
• Determine which cells the start and goal lie in, and search for
a path in the adjacency graph between these cells.
• From the sequence of cells found in the last step, compute a path
connecting certain points of cells such as their midpoints (centroids)
via the midpoints of the boundaries.

11
Vertical strip
cell decomposition

• The cells joining the start


and target are shaded.
• The resolution of the
decomposition is chosen
to get a collision-free path
dependent on the
sensitivity of controlling
robot’s motion.

12
Trapezoidal
Cell Decomposition

• as

13
Grid representation

• For movements in 8 directions, the


scene decomposition Grid
representation is used.
• Robot size must be smaller than cell
size.
• In the opposite case, we are not able
to determine uniquely the robot
position. This decreases the possible
range of grid.
14
Limitations in
Grid representation

• as

15
Grid representation

16
Grid representation

17
Voronoi Diagram
• A Voronoi diagram of a set of
sites in the plane is a
collection of regions that
divide up the plane.
• Each region corresponds to
one of the sites and all the
points in one region are
closer to the site
representing the region than
to any other site.
18
Example
• RATP open data
website, the public
transport operator in
Paris, at
http://data.ratp.fr

19
What is motion planning?

• Robot Configurations

20
What is motion planning?

• Robot Configurations

21
What is motion planning?

• Robot Configurations

22
What is motion planning?

• Simple idea: grid + search

23
What is motion planning?

• Simple idea: grid + search

24
What is motion planning?

• Create a tree of possible paths

25
What is motion planning?

• Create a tree of possible paths

26
What is motion planning?

• Create a tree of possible paths

27
What is motion planning?

• Create a tree of possible paths

28
Idea

• Create a tree of possible paths

29
Idea

• Create a tree of possible paths

30
Idea

• Create a tree of possible paths

31
Idea

• How can we search possible


paths efficiently?

32
BFS

• Breadth-First Search (BFS)


33
BFS

• BFS is a vertex based technique.


• It uses a Queue data structure which follows first in first out.
• In BFS, one vertex is selected at a time when it is visited and marked
then its adjacent are visited and stored in the queue. It is slower
than DFS.
• BFS is more suitable for searching vertices which are closer to the
given source.
• BFS considers all neighbors first and therefore not suitable for
decision making trees used in games or puzzles.
34
DFS

• Depth-First Search (DFS)


35
DFS

• DFS is a edge based technique.


• It uses the Stack data structure, performs two stages, first visited
vertices are pushed into stack and second if there is no vertices
then visited vertices are popped.
• DFS is more suitable when there are solutions away from source.
• DFS is more suitable for game or puzzle problems. We make a
decision, then explore all paths through this decision. And if this
decision leads to win situation, we stop.

36
Uninformed Search Algorithms

• Depth-limited search - a pre-defined limit


up to which it can traverse the nodes.
Depth-limited search solves one of the
drawbacks of DFS as it does not go to an
infinite path.
• Iterative deepening depth-first search -
a combination of DFS and BFS.
• IDDFS find the best depth limit by gradually
adding the limit until the defined goal
37 state is reached.
Other Search Algorithms

• Bidirectional search - It executes two simultaneous


searches called forward-search and
backwards-search and reaches the goal state.

• Uniform cost search -


It searches the graph
by giving maximum
priority to the lowest
cumulative cost.
38
Informed search algorithms

• Greedy best-first search algorithm - Greedy


best-first search traverses the node by
selecting the path which appears best at
the moment. The closest path is selected
by using the heuristic function.
• A* search algorithm - is a
combination of both uniform
cost search and greedy best-first
search algorithms.
39

You might also like