Obstacle Detection and Avoidance for Efficient UAV Path Planning

Obstacle detection and avoidance is becoming the foremost need and point of concern in navigation and control of Small Unmanned Aerial Vehicles (SUAV). SUAV path planning is a challenging task due to variability of environment and different kinds of objects at low attitude in populated area. It is n

Project Title

Obstacle Detection and Avoidance for Efficient UAV Path Planning

Project Area of Specialization

Artificial Intelligence

Project Summary

Obstacle detection and avoidance is becoming the foremost need and point of concern in navigation and control of Small Unmanned Aerial Vehicles (SUAV). SUAV path planning is a challenging task due to variability of environment and different kinds of objects at low attitude in populated area. It is not ensured that the planned path will be exactly same, without any change, and no obstacle. In this project, we propose an obstacle detection and avoidance in SUAV planned path using image-processing. However, the major issues arise while processing aerial images are: Images with cluttered background, images with low resolution and blurry, images with variable lightning conditions and images with environment variability.

The implementation of the project consists of following: (1) Raspberry-PI based platform for image and video acquisitioning. (2) Data link for transmission of video/image from UAV to ground processing system (2) Enhancement of images and pre-processing steps (3) Extraction of region of interest (4) Objects detection (5) Recognition of objects as obstacles (6) Obstacle modeling and eventually obstacle avoidance (7) Real-time to UAV to avoid the obstacle.  In this project, both MATLAB and Python are used for implementation of Obstacle Detection and Avoidance for Efficient UAV Path Planning. The dataset of 20k images is taken. We will evaluate performance comparisons of various algorithms based on efficient energy consumption, shortest alternate path and time complexity.

Project Objectives

• SUAV Platform for real-time obstacle detection and avoidance using Raspberry-PI system-on-chip
• Image processing algorithms and techniques for obstacle detection and avoidance
• Modelling of obstacles for avoidance.
• Collection of various images and distance data set using system installed.
• Performance comparisons of various modeling algorithms.   

       

Figure 1. Objective Diagram for Obstacle Detection and Avoidance for Efficient UAV Path Planning

Project Implementation Method

                

Figure 2. Process Overview Diagram for Obstacle Detection and Avoidance for Efficient UAV Path Planning

Figure 3. Implemented Architectural Diagram for Obstacle Detection and Avoidance for Efficient UAV Path Planning

          

Figure 4. Flowchart  Diagram for Obstacle Detection and Avoidance for Efficient UAV Path Planning

Benefits of the Project

• Unmanned Aerial Vehicles (UAVs) are applied on several applications like mapping, journalism, transport, rescue military applications and environments where a human cannot access.
• UAVs have a great potential to replace manned aircraft in many tasks or missions e.g. surveillance, search and rescue missions, attack missions and collecting data etc.
• This Project can be further utilized for:
  • Saving lives.
  • Supporting law enforcement.
  • Contributing to safe infrastructure, maintenance and management.
  • Streamlining agricultural management.
  • Giving media access to hard-to-reach places.

Technical Details of Final Deliverable

• Technology development for UAV platform and on-board platform for images acquisitioning APIs

         Still Picture Resolution: 2592 x 1944

         Video: Supports 1080p @ 30fps, 720p @ 60fps and 640x480p 60/90 Recording

• Technology development for establishing gateway/Hardware datalink between on-board raspberry Pi and ground processing systems during UAV’s flight

           Mavlink

• Software technological developmental algorithm for detection of obstacles in aerial images

           Viola-jones

           Template matching

• UAV and Raspberry PI based system for obstacle detection and avoidance

           Waypoints tracking

• Theoretical analysis and technical reports of implemented systems

          Finalized achieved results in forms of tables, graphs and charts

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Transportation

Other Industries

Security

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

Industry, Innovation and Infrastructure, Responsible Consumption and Production

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com