Skip to content

All Hands On Deck : Autonomous Underwater Robot With AI Made By A Team Of Students.

All Hands On Deck : Autonomous Underwater Robot With AI Made By A Team Of Students.

horizontal line

What is an AI Powered Underwater Robot ?

Underwater robot is a device which is designed and programmed to go deep in water for exploration and for gaining knowledge. They are able to go to remote as well as dangerous places where humans cannot go by themselves or if it is too risky for humans to go to that place. It can also go to some unexplored places of the ocean so as to explore its characteristics. It can measure temperature, salinity, speed and direction of currents, etc. It is also used to examine and observe some underwater fishes and creatures that are deep inside the ocean. This is known as an underwater robot.

Now let us know the insight of a competition where a team of students made an AI powered underwater robot.

Insight of the competition :

There was an annual competition in San Diego. In it there were many student teams from around the world that took part in the competition to design and build AI powered submarines.

The competition consisted of a timed obstacle by using real time tasks that were designed to test the AUV i.e. Autonomous Underwater Vehicle. It tested which vehicle was able to do such things like navigation in the underwater gates, mapping of buoys, launching some torpedo shaped markers, etc. The above mentioned tasks require AUVs to have swift movements and precise navigation systems. For all these activities a vehicle needs to sense its environment to give precise answers.

The name of the team is SONIA. It is based in Montreal, Quebec and has competed in many competitions with over 20 students in them. The name of the captain and the mastermind is Martin. He believed that while equipping AUV with sensors, precision is very important.

Technology Involved In The AUV :

The technology that is used in the underwater vehicle that they made is as follows :

  • Two cameras.
  • A Doppler Velocity Log ( DVL ).
  • IMU.
  • 4 hydrophones.
  • Mechanical Imaging Sonar.

The sensors equipped in the underwater vehicle helps the vehicle to see, hear, and measure the speed, acceleration and the position of the AUV. It was also able to measure the accurate distance of the objects that are ahead of it. To interact with its environment the AUV is well equipped with 6 thrushers, grabbing system, and a torpedo launcher.

These equipments used computer vision.

What Is Computer Vision ?

Computer vision in simple terms is a computer based system for visualising its surroundings and   helps computers to derive meaningful information from digital images, videos and other types of  visual inputs. It takes actions according to the retrieved information. If AI helps computers to think on its own, then, computer vision helps computers to visualise, observe and understand its surroundings. This is known as computer vision.

The main intention of integration of Computer Vision (CV) in Artificial Intelligence is to make and perform visual perception that can visualize and analyse the situation and perform the task on its own without the help of humans. This complete procedure involves certain methods like, obtaining datasets, processing and analyzing them, and also understanding the digital images to make use of them effectively  in the real-world.

Computer vision in Artificial Intelligence plays an important role in developing Machine Learning models for multiple sectors.Process involves from object detection to expression recognition, Computer Vision is providing detailed information about multiple things to gadgets around the world.

Mission of the AUV :

The mission of the AUV is to demonstrate autonomy by fulfilling the games. SONIA already had enough training data their model was detecting objects quite well without taking any data from the organisers of the competition. 

Now, below is a shortlist of their ML work :

  • Extracting images from the raw data obtained from ROS i.e. robotic operating system.
  • Transforming extracted images into tensorflow records format.
  • Train and tune hyperparameters.
  • Train the model using API. Training time is almost 5 hours per model.

Now that you have reached this far, I assume that  you liked the blog and you have got some value from it. To continue embracing yourself with similar knowledge, just click on the link below:

Something About Ourselves :

24×7 offshoring is an IT, data, and business process outsourcing firm situated in Delhi, India. 24x7offshoring is the one-stop solution for multinational organizations all over the world, with experience in over 100 media to large-scale projects across five continents.

Dr. Teja is the company’s co-founder, and also the CEO in charge. He has a lot of expertise in managing, coordinating, and delivering high-profile initiatives. We are confident in world-class project management and delivery across domains because of the network of experts and clients we have developed over the previous three years.

Our motto is “Our network is our net worth.” Our long-term relationships with all of our clients have aided our 300 percent year-over-year growth.

No comment yet, add your voice below!

Add a Comment

Your email address will not be published.

Request for Call Back

Welcome to 24x7Offshoring. Enter your details to contact us.