Suhrudh. S

Suhrudh. S

Undergraduate Student at BITS Pilani, Goa

Heuristically guided Sampling based Path Planning

less than 1 minute read

Published:

Sampling based Path planning algorithms suffer for optimality and convergence issues. Algorithms like RRT, Informed RRT etc. are said to asymptotically optimal; i.e. the algorithm converges to optimal solution as the number of samples tend to infinity. But there is no proper gaurentee given about the rate of convergence. To improve the rate of convergence, algorithms like BIT* perform heuristically guided search to converge faster towards optimalitily.

Here is the link to the blog post I wrote more on this.

I also gave a talk on Sampling based Algorithms at Paper Discussion Sessions as a part of Club activities at the Electronics and Robotics Club. The links to the presentation and the recordings are given below

You May Also Enjoy

Autonomous Drone Navigation

5 minute read

Published:

During the pandemic, I was at home without any access to any physical robots. It was the perfect time to put into practise the concepts of Robotics that I learnt in the summer. I got quite intriguied by Autonomous drones. My greatest inspirations came after watching this video on Deep Drone Racing by the Robotics and Perception group at UZH and a bunch of really cool work by the Autonomous Robots Lab at NTNU. Taking inspiration, I wanted to try out working with Indoor Navigation of Autonomous Drones.

Path Planning in Autonomous Robots

7 minute read

Published:

Path planning is one of the most important primitives for autonomous mobile robots. The ability to be able to travel on its own by finding a collision free, optimal path is an important aspect of making robots autonomous. Path, as the name suggests is a set of waypoints which a Robot is expected to travel. There can be many criterions for deciding a path that the Robot should follow. Various optimisations, checks are made before deciding an optimial path.