@mihirk284 Profile picture

Mihir Kulkarni

@mihirk284

PhD Student, @NTNU, @NTNUNorway | Mechanical Engineering @bitspilanigoa 2020.

Similar User
Aditya Kapoor photo

@_AdityaKapoor_

Pranav Mahajan photo

@iam_mahajan

Mihir Dharmadhikari photo

@MihirDharmadhi1

Vishwa Shah photo

@vishwayvs

Aditya Lahiri photo

@cross_entrippy

Avishree Khare photo

@Avishreekh

Hermann Blum photo

@hermannsblum

Aditya Bidwai photo

@adbidwai

Mehul Rastogi photo

@mehulr_98

Shrey Pandit photo

@ShreyPandit2001

Sharad Chitlangia photo

@sharad24sc

Ajay Subramanian photo

@ajaysub110

Vedant Shah photo

@veds_12

Ashwin Vaswani photo

@ashwin_vaswani

Ojit photo

@OjitMehta

Mihir Kulkarni Reposted

We are happy to release a major update for the "Aerial Gym Simulator". * Code:  github.com/ntnu-arl/aeria… * Video: youtu.be/ROQ8iK_l1jM The work was part of the @spear_robotics project funded by @EU_HaDEA and @AFOSR #robotics #simulation #autonomy #deeplearning


Mihir Kulkarni Reposted

As part of the recently open-sourced "Aerial Gym Simulator 2.0", we have included a set of training examples including using reinforcement learning for navigation using depth images & without the need for a map. Instructions: ntnu-arl.github.io/aerial_gym_sim… #robotics #rl @mihirk284


Mihir Kulkarni Reposted

If you are interested in ensuring safety for navigation policies relying on high-dimensional exteroceptive data (e.g., LiDAR, vision), check out our recent paper on *Neural Control Barrier Functions*. Outstanding work from Marvin Harms and team! arxiv.org/abs/2407.19907 #robots


Mihir Kulkarni Reposted

We present *Morphy*, a novel Compliant and Morphologically-aware Flying #Robot that integrates sensorized flexible joints in its arms, thus enabling resilient collisions and the ability to squeeze through openings more narrow than its nominal dimensions. youtu.be/C6l7Vklbc9k?si…


Mihir Kulkarni Reposted

@mihirk284 presenting our work on Reinforcement Learning for Collision-Free Flight Exploiting Deep Collision Encoding @ieee_ras_icra

Tweet Image 1

Mihir Kulkarni Reposted

Short summary video for our @ieee_ras_icra paper on Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding while new release of our Aerial Gym simulator (based on Isaac) is coming soon! youtu.be/IpGAmW7ZevY #robotics #reinforcementlearning #drones


Mihir Kulkarni Reposted

Deadline closing soon (March 12, 2024): Exciting new opportunity for a PhD on Multi-modal Long Horizon Navigation Learning for Aerial Robots #robotics #ntnu #phd jobbnorge.no/en/available-j…


Great stuff @ki_ki_ki1 !!

We trained ANYmal to go into confined spaces such as under collapsed buildings. To be presented at #ICRA2024 Title: Learning to walk in confined spaces using 3D representation Arxiv: arxiv.org/abs/2403.00187 Video: youtu.be/QAwBoN55p9I Summary Page: takahiromiki.com/publication-po…



Mihir Kulkarni Reposted

Mihir Kulkarni, Kostas Alexis, "Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding", IEEE International Conference on Robotics and Automation, May 13-17, Yokohama, Japan youtu.be/gPrT21sbpTY


Mihir Kulkarni Reposted

Excited for our 4 #ICRA2024 papers: *Online Self-calibrating Refractive Camera Model with Appl. to Underwater Odometry *Degradation Resilient LiDAR-Radar-Inertial Odometry *RL for Collision-free Flight Exploiting Deep Collision Encoding *N-MPC for DNN-Based Collision Avoidance

Tweet Image 1

Mihir Kulkarni Reposted

We are excited to *open-source* the ORACLE library of methods for #deeplearning-based safe navigation of aerial #robots that assume a) no access to a map of the environment or an estimate of the robot’s position, & present robust sim2real transfer. Code: github.com/ntnu-arl/ORACLE


Excited to share the results of our latest work! We propose an RL-based low-latency, map-free navigation method that exploits a compressed representation of depth images and generates action commands to navigate cluttered environments.

Do you care for RL navigation of drones without many assumptions for the type of the environment and without the use of online mapping? Check the results of our work on Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding youtu.be/gPrT21sbpTY?si…



Loading...

Something went wrong.


Something went wrong.