chahatdeep / rrt_dynamic

This is an RRT demonstartion for a finite volume robot with kinodynamic constraints.

RRT Smart Algorithm for a finite volume robot:

This is an RRT implementation for a finite volume robot with kinodynamic constraints.

Rapidly-Exploring Random Tree

Author: Chahat Deep Singh

April 25 2017

University of Maryland College Park

---

Instructions:

• Run the file RunMe.m in Matlab (R2009b+)

• Make sure the files: (a). CollisionFree (b). H5_robot.txt (c). Near (d). Nearest (e). ObstacleFree (f). RRTs (g). Steer (h). Steer2 (i). circle (j). circleExtrude (k) confun (l) mat_file_cal (Not required by RunMe; Needed for CSV file generation) (m) plotObstacle (n) plotObstacleExtrude (o) robot are in the same folder.

• mat_file_cal creates CSV files which contains the time, position (x,y), theta, velocity, omega, acceleration and gamma (steer acc) in the respective order.

Input Arguments:

• xdim, ydim : Size of the grid
• Start Node: start_node [x y]
• Target: goal [x y]
• radius: Radius of the End Goal region
• r: radius of the obstacles
• Epsilon: Minimum Step size for the RRT
• Change the input data here

Sample Outputs

• Output 1
• Output 2
• Output 3
• Output 4
• Output 5

Sample CSV Files

• CSV 1
• CSV 2
• CSV 3
• CSV 4
• CSV 5

Requirements

• Matlab R2009b or later.

Notes

• Tested on MATLAB R2016b, Arch Linux 2017.03.01, Kernel: 4.9.11