Internship: Robotics Engineer (5)

EasyMile is hiring!

About

Pioneer in driverless mobility, EasyMile revolutionizes passenger and goods transportation.
For the past years, EasyMile has developed and deployed autonomous mobility solutions worldwide based on vehicles manufactured by recognized industrial partners. EasyMile’s clients include the world’s largest transport operators, city authorities, airports, corporations, business parks, and universities.
EasyMile supplies smart mobility solutions and autonomous technologies powering driverless vehicles with existing products such as: the EZ10, a 100% electric driverless shuttle deployed in 20 countries worldwide; the TractEasy, the first autonomous electric baggage tractor jointly developed with TLD; and many other projects to be unveiled soon.

EasyMile’s team is composed by more than 230 high skilled and passionate employees specializing in robotics, computer vision and vehicle dynamics.
The company has its headquarter in Toulouse (France) and regional offices in Singapore (Asia), Denver (USA), Berlin (Germany) and Adelaide (Australia).

Job Description

We are looking for highly motivated people to build the road ahead and invent solutions to challenges never met before. Creativity and perseverance is your attitude? We are looking for you.

  • Duration: 6 months (starting from March 2021)
  • Location: Toulouse, France
  • Department: R&D - Navigation
  • Internship subject: Study agent motion prediction methods
  • Internship tutor: Axel Lefrant, R&D Enigineer

The goal of this internship is to study the techniques for adverse agents movement prediction (agents refer to the dynamic objects detected by the tracking modules, like vehicle, pedestrian, etc.) and realize a benchmark study of the promising techniques to be tested on simulated and real world data produced by our EZ10 autonomous shuttle.

The internship will consist of different phases:

Study state of the art

  • Classification of the techniques (using physics model, motion planning approach, agents cues, environment cues, etc.).
  • See for instance Rudenko, Andrey, et al. "Human motion trajectory prediction: A survey." The International Journal of Robotics Research 39.8 (2020): 895-935.
  • Define for each technique the inputs (i.e. cues) and the outputs (i.e. reachable set, path, particles, etc.)
  • Evaluate the time horizon and the computational constraints

Write scenarios and metrics to test and compare the prediction techniques

  • Elaborate rational criteria to sort techniques in term of accuracy, computational cost, efficiency and depending on the driving scenario.

Experiment on simulated and/or real data set (Python and/or C++)

  • Implement promising prediction techniques
  • Compare the prediction output with the agents ground truth
  • Evaluate performances on simulated dataset, on the EasyMile simulation testbench and on real data recorded on our vehicles

Study the connexion with the decisions made by the navigation software (closed loop)

  • What are the impacts of the prediction on the navigation
  • Test on simulated environment and possibly on real vehicle

Preferred Experience

There is no typical profile at EasyMile, we all come from different backgrounds and that is what makes us strong! Don't hesitate to apply if you are motivated and interested by innovative transportation and technologies.

We are looking for:

  • Final year internship
  • C++
  • Python
  • Strong mathematical background
  • Nice to have: Robotics experience with ROS, estimation, kinematic and/or dynamic modelling

Recruitment Process

  • 30 minutes call with a recruitment officer
  • 30 minutes call with the tutor
  • One hour interview with the tutor and a recruitment officer

All our offers are open to people with disabilities

Additional Information

  • Contract Type: Internship (Between 6 and 6 months)
  • Location: Toulouse, France (31000)
  • Education Level: Master's Degree