https://github.com/praveenkumar-rajendran/udacity-sensor-fusion-nanodegree

Projects Implemented for the Udacity Sensor Fusion Engineer Nanodegree Program
https://github.com/praveenkumar-rajendran/udacity-sensor-fusion-nanodegree

3d-object-tracking autonomous-vehicles cfar computer-vision descriptor-extraction euclidean-clustering feature-matching fmcw-waveform kd-tree keypoint-detection lidar object-detection obstacle-detection radar-detection ransac sensor-fusion time-to-collision ttc-computation udacity-nanodegree unscented-kalman-filter

Last synced: 3 months ago
JSON representation

Projects Implemented for the Udacity Sensor Fusion Engineer Nanodegree Program

• Host: GitHub
• URL: https://github.com/praveenkumar-rajendran/udacity-sensor-fusion-nanodegree
• Owner: PraveenKumar-Rajendran
• Created: 2024-06-23T08:28:13.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-06-23T10:50:41.000Z (12 months ago)
• Last Synced: 2025-02-01T05:42:16.582Z (5 months ago)
• Topics: 3d-object-tracking, autonomous-vehicles, cfar, computer-vision, descriptor-extraction, euclidean-clustering, feature-matching, fmcw-waveform, kd-tree, keypoint-detection, lidar, object-detection, obstacle-detection, radar-detection, ransac, sensor-fusion, time-to-collision, ttc-computation, udacity-nanodegree, unscented-kalman-filter
• Language: C++
• Homepage: https://www.udacity.com/course/sensor-fusion-engineer-nanodegree--nd313
• Size: 72.8 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Udacity Sensor Fusion Nanodegree Program Projects

## Projects

### 01. [SFND Lidar Obstacle Detection](01_SFND_Lidar_Obstacle_Detection)

![Obstacle Detection](https://video.udacity-data.com/topher/2019/March/5c8599e3_obstacledetectionfps/obstacledetectionfps.gif)

  Description

    Implementation of custom RANSAC, KD-Tree, and Euclidean clustering algorithms as part of the processing pipeline for Lidar obstacle detection.

### 02. [SFND 2D Feature Matching](02_SFND_2D_Feature_Matching)

![Keypoints](02_SFND_2D_Feature_Matching/images/keypoints.png)

  Description

    Implementation of various detectors, descriptors, and matching algorithms. It consists of four parts: data buffer, keypoint detection, descriptor extraction and matching, and performance evaluation.

### 03. [SFND 3D Object Tracking](03_SFND_3D_Object_Tracking)

![3D Object Tracking](03_SFND_3D_Object_Tracking/assets/Fast-orb-perfect.png)

  Description

    Implementation of the following components:

    - Matching 3D objects

    - Computing Lidar-based TTC

    - Associating keypoint correspondences with bounding boxes

    - Computing Camera-based TTC

    - Performance evaluation

### 04. [Radar Target Generation and Detection](04_Radar_target_generation_and_detection)

![Radar Target Generation and Detection](04_Radar_target_generation_and_detection/assets/01_project_layout.png)

  Description

    Implementation of radar target generation and detection:

    - FMCW waveform design

    - Simulation loop

    - Range FFT (1st FFT)

    - 2D CFAR

### 05. [SFND Unscented Kalman Filter](05_SFND_Unscented_Kalman_Filter)

![Unscented Kalman Filter](05_SFND_Unscented_Kalman_Filter/media/ukf_highway.png)

  Description

  The simulation collects the position and velocity values output by the algorithm and compares them to the ground truth data. The px, py, vx, and vy RMSE values have been implemented to be less than or equal to [0.30, 0.16, 0.95, 0.70] after the simulator runs for more than 1 second. The simulator also displays if the RMSE values exceed the threshold.

## Udacity - Graduation Certificate

![Obstacle Detection](assets/SensorFusion-Nanodegree.png)