AUTODRI

Join us for the closing webinar where top participants will discuss their solutions!  

🛠 Contribute: Found a typo? Or any other change in the description that you would like to see? Please consider sending us a pull request in the public repo of the challenge here.

🕵️ Introduction

Did you know, we have been experimenting with driverless cars at least since the 1920s? 🤯 For reference, that was the time when Ford pioneered mass-market automobile manufacturing and still produced the Model T!

The human need for removing humans out of every repetitive task is astonishing. With this challenge, we take that need even further.

In this challenge, we give you images of four camera angles ( front,left,right,rear) and you have to use any subsut of these image to predict the CAN steering angle of the next second for safe driving!

Creating better autonomous driving solutions to the point that we can implement them, will systematically help tackle road accidents and eliminate them. With this challenge, we are proud to initiate efforts to tackle SDG 3.6 - road traffic accidents.

Understand with code! Here is getting started code for you.😄

💾 Dataset

The dataset contains images of 4 camera angles(front, left, right, rear ) for each image id i.e. for each position of a self driving car. It also contains a csv containing the steering wheel angle for the next second in future for safe driving. One can use any subset of these camera angles to predict the steering wheel angle. The range of the CAN steering angle is from [-720,720].

📁 Files

Following files are available in the resources section:

• train.zip - (44304 samples) This folder contain 4 subfolders cameraFront.cameraLeft, cameraRight, cameraRear and 1 csv train.csv. The subfolders contain the images of car in the respective angle and the csv contains the image id and its steering angle,

• test.zip - (21269 samples) This folder contain 4 subfolders cameraFront.cameraLeft, cameraRight, cameraRear and 1 csv train.csv. The subfolders contain the images of car in the respective angles and the csv contains the image id without the can steeting angle,

• val.zip - This folder contain 4 subfolders cameraFront.cameraLeft, cameraRight, cameraRear and 1 csv train.csv. The subfolders contain the images of car in the respective angle and the csv contains the image id and its steering angle,

• sample_submission.csv - Sample submission format for the challenge.

🚀 Submission

• Prepare a CSV containing header as [filename,canSteering], where filename is the name of the image and canSteering is the predicted value of Controller Area Network for the image.
• Name of the above file should be submission.csv.
• Sample submission format available at sample_submission.csv in the resorces section.

Make your first submission here 🚀 !!

🖊 Evaluation Criteria

During evaluation Mean Squared Error be used to test the overall performance of your solution.

🔗 Links

• 💪 Challenge Page: https://www.aicrowd.com/challenges/autodri
• 🗣️ Discussion Forum: https://www.aicrowd.com/challenges/autodri/discussion
• 🏆 Leaderboard: https://www.aicrowd.com/challenges/autodri/leaderboards

📱 Contact

• Rohit Midha
• Shraddha Mohan
• Sharada Mohanty
• Ayush Shivani