ennasachdeva Goto Github PK

2r-manipulator-force-control

This is a MATLAB simulation for force control of a 2R manipulator using feedback linearization.

algorithms

cifar-10-binary-classification

computer-vision-chroma-keying

convert_kitti_to_ros

A useful ROS tool for dealing with KITTI point cloud dataset.

cppnd-route-planning-project

cvpr_dnri

Code accompanying "Dynamic Neural Relational Inference" from CVPR 2020

d_vae

deep-learning-implemetations

ekf-localization

ennasachdeva.github.io

Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes

image-mosaicing

iros-2017-cocrip-optimization

Optimization for estimating friction coefficient of materials to be used for In-Pipe climbing robot COCRIP (published in IROS-2017) for vertical and bend pipes.

leisure_time_stuff

machine_learning_algos

Machine learning algorithms

madys

Code accompanying Multiagent Learning via Dynamic Skill Selection

maedys

multiagent_erl_heterogeneous_rover_domain

non-holonomic-trajectory-planning-using-the-bernstein-basis-functions

papers-summary

This repository includes the summary of various papers in Robotics and AI, I have read.

recurrent-multiagent-deep-deterministic-policy-gradient-with-difference-rewards

Deep Reinforcement Learning (DRL) algorithms have been successfully applied to a range of challenging simulated continuous control single agent tasks. These methods have further been extended to multiagent domains in cooperative, competitive or mixed environments. This paper primarily focuses on multiagent cooperative settings which can be modeled for several real world problems such as coordination of autonomous vehicles and warehouse robots. However, these systems suffer from several challenges such as, structural credit assignment and partial observability. In this paper, we propose Recurrent Multiagent Deep Deterministic Policy Gradient (RMADDPG) algorithm which extends Multiagent Deep Determinisitic Policy Gradient algorithm - MADDPG \cite{lowe2017multi} by using a recurrent neural network for the actor policy. This helps to address partial observability by maintaining a sequence of past observations which networks learn to preserve in order to solve the POMDP. In addition, we use reward shaping through difference rewards to address structural credit assignment in a partially observed environment. We evaluate the performance of MADDPG and R-MADDPG with and without reward shaping in a Multiagent Particle Environment. We further show that reward shaped RMADDPG outperforms the baseline algorithm MADDPG in a partially observable environmental setting.

reinforcement-learning_arthur-s-bar-problem

releasing-research-code

Tips for releasing research code in Machine Learning (with official NeurIPS 2020 recommendations)

research-papers

Research papers on Robotics, Deep learning, Reinforcement learning

resume-screener

A project for CodeDay Labs that screens resumes based on their fit for a Software Engineer New Grad position

robotics-path-planning

robotics_informative-path-planning

rover-domain

scenario_runner

Traffic scenario definition and execution engine

scribe-notes

Scribe notes on various topics relevant to robotics and AI.