That is correct. It is possible to update the obsatcles at real time using other sensors like LiDAR and cameras, but that was not done in this project.
Yup, the robot is able to move autonomously from one point to another while avoiding all obstacles. The code has two main components: path generation and path following. Path generation generates a path based on the target point that avoids all the marked obstacles. My code uses an algorithm called Rapidly exploring Random Tree* (RRT*) to generate the path. Path following refers to making the robot follow the set path accurately. I made the robot drive toward a point on the path that is x units away from the robot. As the robot moves forward, the target point gets pushed back and the robot progressively follows the path. You can see a simulation of what is happening in this video below. czcams.com/video/MyCptpi5Ie4/video.html&ab_channel=RyanLiao-4253B Path generation generates a path based on the target point that avoids all the marked obstacles. My code uses an algorithm called Rapidly exploring Random Tree* (RRT*) to generate the path.
Nice
I’m assuming the obstacles positions are pre determined? And then in path generation it just avoids points that would collide
That is correct.
It is possible to update the obsatcles at real time using other sensors like LiDAR and cameras, but that was not done in this project.
How you do that? I think that its autonomous, right?
Yup, the robot is able to move autonomously from one point to another while avoiding all obstacles. The code has two main components: path generation and path following.
Path generation generates a path based on the target point that avoids all the marked obstacles. My code uses an algorithm called Rapidly exploring Random Tree* (RRT*) to generate the path.
Path following refers to making the robot follow the set path accurately. I made the robot drive toward a point on the path that is x units away from the robot. As the robot moves forward, the target point gets pushed back and the robot progressively follows the path.
You can see a simulation of what is happening in this video below.
czcams.com/video/MyCptpi5Ie4/video.html&ab_channel=RyanLiao-4253B
Path generation generates a path based on the target point that avoids all the marked obstacles. My code uses an algorithm called Rapidly exploring Random Tree* (RRT*) to generate the path.