2022 Build-A-Thon

ITU India AI/ML Challenge

Low Latency Closed Loop
Winners Announced for Build-a-Thon. Click here to view
About us
  • Indian Institute of Technology Delhi: Technical and Management Team
  • Telecommunications Standards Development Society, India: Supporting Partner
  • Ministry of Electronics & Information Technology: Technical Partner
  • International Telecommunication Union: Technical partner
  • The India-EU ICT Standardization Collabaration Project is funding and supporting the Build-a-Thon with Experts from EU and award for the winning Team.

Overview

Overview
To provide a platform for students and professionals in India to participate in global pre-standards activity and Build-A-Thon challenge related to AI/ML in 5G. IIT Delhi with support from India EU ICT Standardization Project, ITU and TSDSI ,launched the India specific Build-A-Thon aligned with ITU global challenge.
  • Build upon the previous work done by IIT Delhi in the area of focus group machine learning in 5G and extend for solving practical problems in the network and society.
  • Proof of concept development, culminating in demos.
  • Collaboratively create knowledge base and publish in global publications.
  • Prepare Final technical report.
  • A knowledge dissemination summary and technical report would be submitted as technical.
  • Report to SDOs to be treated as pre-standards input. This can lead to further standard proposals after due study and gap analysis of existing standards.
Prizes:
  • The winning team will get opportunity to enter the global ITU 2022 Build-A-Thon challenge.
  • Study tour to EU for winning team.
  • Winning team (with most practical design) will be invited to IITD for testing their solution on actual hardware.

Problem Statement

Problem statement I : Slip Detection (and Force Estimation)

  • Given: Object being held by Robotic Hand.
  • The object may tend to slip from the grip (by introducing weight change to object).
  • This task will have constant simulated real world constraints like gravity.
  • Overall this task requires an estimation of the force to be applied on the object for a successful grasp.
  • It is desirable to leverage latency-aware learning technique to
    •   Find if the object is about to slip
    •   Applying the appropriate control to prevent the slip
      •     Without breaking or deforming the object (force estimation)
      • Time series Labeled Datasets will be provided to the registered teams by IITD team.

        Problem statement II : Object Detection

        • Given: Object currently held by the Robotic Hand.
        • The Robotic Hand records its current states (in the form of angular configuration and forces exerted on its joints)
        • Using this modality the task is to detect the shape of the object.
        Time series Labeled Datasets will be provided to the registered teams by IITD team.
    About Us

    IIT Delhi has developed a MEC Test Bed integrated with 5G Core conforming ETSI standards (https://bhartischool.iitd.ac.in/mec-5g/). The application developed by the selected contestants will be tested on the Test-Bed with actual hardware (Allegro Hand and Haptic glove). All the required APIs for the test bed are already there and Test-Bed Team will help the contestants to integrate it with the Test-Bed.

    Announcemets

    Events & Announcements

    Winners:

    # Team Name Position
    1 Roger That Winner
    2 AI ML SSD Ist Runners Up
    3 TCS Smart Machine IInd Runners Up

    Important Dates:

    # Date Mode of Event Event Title
    1 12th Sep, 2022 Online Round Table (IITD)
    2 23rd Sep, 2022 Hybrid(Online/Offline) Round Table 2 (IITD)
    2 Deadline: 30th Sep, 2022 Online Pre-evaluation of Problem Statement II (IITD)
    2 Deadline: 3rd Oct, 2022 Online Pre-evaluation of Problem Statement I (IITD)
    2 11th Oct, 2022 Online Final Submissions of Solution (IITD)
    3 14th Oct, 2022 Online Declaration of Results (IITD)
    4 14th Oct, 2022 Hybrid Build-a-thon Workshop, with invited talks from EU Experts (IITD)
    5 28th Oct, 2022 Online Start of Global round (ITU)
    6 25th Nov, 2022 Online Announcement of Global Awards (ITU)
    7 Dec, 2022 EU trip to winners

    Past Events
      Pre-Evaluation


        Problem statement I - Slip Dtetection and Force Estimation
        • Training Dataset Download
        • Sample Dataset download
          1. INSTRUCTIONS FOR PRE-EVALUATION OF PROBLEM STATEMENT I: DATASET AND SUBMISSION
          2. You are given a pre-evaluation dataset for the Force Estimation Task. Click on the link to download the pre-evaluation dataset https://drive.google.com/file/d/1O0Ro9fV39rG1facUrLaFtckgZ2KYsFtY/view?usp=sharing. The pre-evaluation dataset is similar to the training dataset but without any labels.
          3. As per the requirements of the task, the input exists in multiple CSV files each relating to an individual time-series of an event of picking up an object of interest that can be deformable and slipping while being picked up.
          4. Parse the pre-evaluation dataset (all the CSV files) and use your model to generate output predictions. These predictions need to be dumped in CSV format such that for each CSV input, there is an output prediction as a CSV file (with exactly the number of rows as input). The name of the output prediction CSV file generated must be the same as that of the input CSV file. Each line in the CSV file contains two labels (one for prediction of slip and one for prediction of crumple, separated with a comma ',' ) corresponding to features in each row in the input CSV. The two labels are separated with a comma (",") i.e., the CSV must contain two column names as "Slip" and "Crumple". Examples of such output prediction CSV files (named as "167552873.287.csv", "1716577643.287.csv") are linked for your reference.
          5. For pre-evaluation, you need to submit only the output predictions folder containing all CSV files (not the code) enclosed in a proper structure defined below.
          6. Your submission must be a ZIP file named as the name of your group (without any spaces, or symbols). For example if your group name is "Babbage Group", the ZIP file should be named as "BabbageGroup.zip".
          7. Inside this ZIP, there should be a folder named as the name of your group (without any spaces, or symbols). For example if your group name is "Babbage Group", the folder name inside the ZIP file should be named as "BabbageGroup". This folder must contain another folder named exactly as "PREDICTIONS_PS1" which contains all output predictions CSV files (whose format is defined in point 2 above).
          8. For an illustration of what your submission should look like, an example ZIP file is linked (named as "BabbageGroup.zip").
          9. For all who need further help to make such a structure of submission, here is what to do: Read an input CSV file, generate predictions from your model against each row of the input into a CSV file. There are two predictions to make for each row in input (viz., for slip and for crumple, separated with a comma). Name this predicted CSV file exactly the same as the input CSV file. Do this for all input CSV files. Create a folder named "PREDICTIONS_PS1". Place ALL the resultant CSV files (just generated) into this folder. Place this folder into another folder with your group/team name. ZIP this folder and name this ZIP file with your Group/Team name (without any spaces or symbols).
          10. Share the ZIP file with us over the mail: mec5giitd@gmail.com with the subject line as "SUBMISSION_PS1".
          11. ONLY use one of the registered emails to send this submission. Failure to send email from at least one of the registered emails will lead to rejection of that submission.
          12. You can send as many submissions as you want to, until the deadline. The latest submission marked with your team name will be treated as final.
          13. The due deadline is 3rd October, 2022, 11:59:59 pm Indian Standard Time.



          Problem statement II - Object Detection
          • Training Dataset download
          • Sample Dataset download
          • INSTRUCTIONS FOR PRE-EVALUATION OF PROBLEM STATEMENT II: DATASET AND SUBMISSION
            1. You are given a pre-evaluation dataset for the Object Detection Task. Pl click on the link to download the dataset https://drive.google.com/file/d/1Uzg0v91V3xdBK4qk-NBQxykUVVf-OaRb/view?usp=sharing . The pre-evaluation dataset is similar to the training dataset but without any labels.
            2. Parse the pre-evaluation dataset and use your model to generate output predictions. These predictions need to be dumped in CSV format such that each line in the CSV file contains a label of the object corresponding to features in each row in the pre-evaluation dataset. An example of such an output prediction CSV file (named as "prediction_output_PS2.csv") is linked for your reference. The first line in the CSV must exactly be the words "Object_Held".
            3. For pre-evaluation, you need to submit only the output predictions CSV file (not the code) enclosed in a proper structure defined below.
            4. Your submission must be a ZIP file named as the name of your group (without any spaces, or symbols). For example if your group name is "Bayesian Group", the ZIP file should be named as "BayesianGroup.zip"
            5. Inside this ZIP, there should be a folder named as the name of your group (without any spaces, or symbols). For example if your group name is "Bayesian Group", the folder name inside the ZIP file should be named as "BayesianGroup". This folder must contain another folder named exactly as "PREDICTIONS_PS2" which contains the output predictions as a single CSV file (whose format is defined in point 2 above). The CSV file can be named with the name of your choice.
            6. For an illustration of what your submission should look like, an example ZIP file is linked (named as "BayesianGroup.zip").
            7. For all who need further help to make such a structure of submission, here is what to do: Generate predictions from your model against each row of the pre-evaluation dataset in an output CSV file. Name this CSV with any name of your choice. As an example, the output prediction CSV file will look like the file linked here with the name "prediction_output_PS2.csv". Create a folder named "PREDICTIONS_PS2". Place the resultant CSV file (just generated) into this folder. Place this folder that contains the CSV into another folder with your group/team name. ZIP this folder and name this ZIP file with your Group/Team name (without any spaces or symbols).
            8. Share the ZIP file with us over the mail: mec5giitd@gmail.com with the subject line as "SUBMISSION_PS2".
            9. ONLY use one of the registered emails to send this submission. Failure to send email from at least one of the registered emails will lead to rejection of that submission.
            10. You can send as many submissions as you want to, until the deadline. The latest submission marked with your team name will be treated final.
            11. The due deadline is 30th September, 2022 up to 11:59:59 pm Indian Standard Time.


          Final Submission Instructions:
          Example file for final submission: TuringGroup.zip


          Important links
          https://github.com/vrra/FGAN-Build-a-thon-2022


        Top Three Leader Board Positions for Pre-evaluation

        Leader board for pre-evaluation is presented here on the basis of "Accuracy" achieved and for any late submission 10% mark is deducted of the original score

        Problem Statement I: Slip detection and Force Estimation

        # Team Position
        1 TCS Smart Machine Ist
        2 AI ML SSD IInd
        3 Roger That IIIrd

        Problem Statement II: Object Detection

        # Team Position
        1 Roger That Ist
        2 AI ML SSD IInd
        2 TCS Smart Machine IIIrd

    CONTACT US

    Contact us and we'll get back to you.

    Prof. Brejesh Lall, Professor, Dept. of Electrical Engineering, IIT Delhi

    Prof. Arzad Alam Kherani, Associate Professor, Dept. of Electrical Engineering and Computer Science, IIT Bhilai

    email: 5g.bhartischool@gmail.com

    email: mec5giitd@gmail.com

    phone: +91 11 26597239

    Copyright © DevCRUD