# reiniscimurs/DRL-robot-navigation

Deep Reinforcement Learning for mobile robot navigation in ROS Gazebo simulator. Using Twin Delayed Deep Deterministic Policy Gradient (TD3) neural network, a robot learns to navigate to a random goal point in a simulated environment while avoiding obstacles.

Repository: https://github.com/reiniscimurs/DRL-robot-navigation
Canonical: https://ross.abutalabs.com/products/drl-robot-navigation
Language: Python
License: MIT
License Family: permissive
Topics: deep-reinforcement-learning, deep-learning, robot-navigation, td3, ros, gazebo, obstacle-avoidance, obstacle-avoidance-robot, reinforcement-learning
Last push: 2025-12-13T21:31:59+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 57, release rhythm 35, longevity 100
- inputs: {"age_days": 1773, "days_push": 263, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1356, forks 193 (observed 2026-08-28T04:04:29.320880+00:00)

## What it is
A ROS Gazebo simulation project that trains a mobile robot to navigate to random goals while avoiding obstacles using a TD3 deep reinforcement learning network. It includes training and testing scripts, a Gazebo environment with a simulated 3D Velodyne lidar, and TensorBoard logging, based on an ICRA 2022 / IEEE RA-L paper.

## Use cases
- train a robot to navigate to goals with deep reinforcement learning
- simulate obstacle avoidance for a mobile robot in Gazebo
- learn TD3 for robot navigation
- reproduce a DRL robot navigation research paper
- set up a ROS Gazebo training environment for navigation policies
- test learned navigation policies with lidar sensor input

## When to choose
- you use ROS Noetic on Ubuntu and want a working DRL navigation setup
- you want to train or evaluate TD3 policies in a Gazebo simulator
- you need a reference implementation for lidar-based obstacle avoidance research

## When to avoid
- you need navigation on real hardware out of the box
- you use ROS 2 or a non-Ubuntu platform
- you want a general-purpose navigation stack rather than a research/training codebase

## Facets
- artifact type: application
- maturity: active
- function: reinforcement-learning, machine-learning, deep-learning, simulation, robotics
- domain: robotics, reinforcement-learning, deep-learning, simulation
- platform: python
- tags: td3, ros, gazebo, obstacle-avoidance, robot-navigation, pytorch, velodyne, linux

## Member repositories
- reiniscimurs/DRL-robot-navigation (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.320880+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:41:55.959608+00:00, confidence not recorded.
  - readme: https://github.com/reiniscimurs/DRL-robot-navigation (fetched 2026-08-28T04:04:29.320880+00:00, sha ddae8e9f25a1)
- Data as of 2026-08-30T08:39:29.467469+00:00.