Presentation is loading. Please wait.

Presentation is loading. Please wait.

Implementation of RRT based Path planner and conversion into Temporal Plan Network By: Aisha Walcott Final Project Presentation Dec. 10, 2001 16.412J.

Similar presentations


Presentation on theme: "Implementation of RRT based Path planner and conversion into Temporal Plan Network By: Aisha Walcott Final Project Presentation Dec. 10, 2001 16.412J."— Presentation transcript:

1 Implementation of RRT based Path planner and conversion into Temporal Plan Network By: Aisha Walcott Final Project Presentation Dec. 10, 2001 16.412J

2 Motivation Scenario: ”Send Aid” Displaced refugees are without food, shelter, and medical supplies Need to send relief aid from another country to their location Problem: Refugees located in a hostile and dangerous region Solution: Send an autonomous air vehicle (UAV) to deliver aid

3 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning: RRT Connect Path Planner Path Planning: RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

4 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning: bRRT Connect Path Planner Path Planning: bRRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

5 Ideally… Design a path planner that adheres to the UAVs dynamic constraints and avoids obstacles Design a path planner that adheres to the UAVs dynamic constraints and avoids obstacles Describe a temporal plan that encodes the “ Send Aid” scenario Describe a temporal plan that encodes the “ Send Aid” scenario

6 Project Objectives Implement a RRT-based path planner Implement a RRT-based path planner Perform analysis based on a set of parameters Perform analysis based on a set of parameters Develop a simulation Develop a simulation Develop an algorithm to transform RRT to a Temporal Plan Network Develop an algorithm to transform RRT to a Temporal Plan Network

7 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning:RRT Connect Path Planner Path Planning:RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

8 Path Planning: RRT Connect Path Planner Grow 2 RRTs into each other Grow 2 RRTs into each other Use a greedy strategy to extend trees Use a greedy strategy to extend trees

9 RRT Connect Path Planner For i = 1 to max_iterations Do 1. q rand = random_config() 2. if ( Extend(T a, q rand ) NOT = Trapped ) then 3. if (Connect( T b, q new ) = Reached ) then 4.Return PATH (T a, T b ) 5. Swap(T a, T b ) Return Failed TaTaTaTa TbTbTbTb q rand Extend: Finds the nearest node in the current tree to q rand Extends a distance  from the nearest neighbor node towards the random node Generates a new node q new if no collision (trapped) Extend(T a, q rand )  Connect: Calls Extend(T b,q new ) until either Extend returns trapped or T a ’s new node was Reached Connect(T b, q new ) q new trapped q near

10 RRT Connect Analysis Input Variables: max_iterations – number of iterations epsilon – extension distance epsilon – extension distance obstacles – arrangement in world obstacles – arrangement in world close_enough – distance of when the 2 RRTs are close enough to be connected Output: path_found = if a path was found or not nodes – total number of nodes generated path nodes – number of nodes along the path path nodes – number of nodes along the path –

11 RRT Connect Analysis Sample Test Case: Sample Test Case: –Vary max_iterations, analyze epsilon values –Output: path found?, total nodes 100,500,1000 iterations: Epsilon = 3,4 no path found Otherwise found path 50 iterations: Found path epsilon> 17

12 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning: RRT Connect Path Planner Path Planning: RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

13 Demo

14 Demo

15 Demo

16 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning: RRT Connect Path Planner Path Planning: RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

17 Temporal Plan Network (TPN) Represents activities Represents activities Represents concurrent activities and decisions between activities Represents concurrent activities and decisions between activities Encodes temporal and symbolic constraints (state of agent or world) Encodes temporal and symbolic constraints (state of agent or world) –Lower bounds, upper bounds –Ask(C) - requires that condition C is true –Tell(C) – asserts the condition C

18 TPN “Send Aid” Scenario: (activities) “Send Aid” Scenario: (activities) –Start UAV engine and load UAV with supplies –Start UAV engine and load UAV with supplies –Take off –Communicate take off status with base –Fly to refugee region by choosing Path A or Path B Path B –Fly to specific location and drop supplies –Fly to specific location and drop supplies –Return to base

19 Temporal Plan Network Drop_Off_Supplies: Fly to refugee region by choosing Path A or Path BFly to refugee region by choosing Path A or Path B Fly to specific location and drop suppliesFly to specific location and drop supplies 3 1 45 8 910 13 2 671112 Drop_Off_Supplies [450,540] Path B [405, 486] Path A Path to Specific Location Drop Supplies [0, 54] [0, 2] [0,0 ] Decision Node Temporal Constraints [ lower_bound, upper_bound ] Activity StartActivity End

20 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning:RRT Connect Path Planner Path Planning:RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

21 Conversion of RRT to TPN Goal: encode the RRT in the TPN Goal: encode the RRT in the TPN Reason: Reason: –A TPN planner exists that outputs a plan for activities that do not involve physical movement –Use TPN planner to plan activities that involve both non-physical and physical movement 8 3 45 67 Path B [405, 486] [0, 0] Path A 8 3 45 67 Path B [405, 486] [0, 0] Path A

22 RRT to TPN RRT After RRT- Connect Planner A E B C D 1. Create two TPN nodes to denote the start and end of the RRT activity 2. If an RRT node n has exactly one child then create a n.start and n.end TPN node 3. Else if RRT node n has more than one child then -go to step 2. -create a TPN decision node -for each child of n go to step 2. C.start C.end E.start E.end A.start A.end B.start B.end D.start D.end Result: RRT becomes a TPN with “dead end” branches

23 Outline Motivation (“Send Aid” scenario) Motivation (“Send Aid” scenario) Objectives Objectives Path Planning:RRT Connect Path Planner Path Planning:RRT Connect Path Planner Demo Demo Temporal Plan Network (TPN) Temporal Plan Network (TPN) Conversion from RRT to TPN Conversion from RRT to TPN Conclusions Conclusions

24 Conclusions RRT Connect Path Planner –Adapts well to problems in non-cluttered environments –Greedy heuristic enables rapid convergence to solution –Max_iterations and extension distance epsilon RRT to TPN –exploits the backtracking of the planner –TPN size >= 2* size(RRT) = O(RRT) –Can express both RRTs and activities in TPN

25 Future Work Current research enables multiple agent representation in TPN Current research enables multiple agent representation in TPN –Rewrite the scenario to include multiple vehicles –Cooperative vehicles Associate a cost with each edge to find optimal plan Associate a cost with each edge to find optimal plan

26 RRT Connect Drawback Demo


Download ppt "Implementation of RRT based Path planner and conversion into Temporal Plan Network By: Aisha Walcott Final Project Presentation Dec. 10, 2001 16.412J."

Similar presentations


Ads by Google