Comparison and benchmarking of the state-of-the-art feature detectors and descriptors for object recognition.
Enhancement of the current object recognition algorithm by the inclusion of alternative visual cues such as color or shape.
Development of clustering techniques for efficient storage and retrieval of feature points.
Evaluation and implementation of the state-of-the-art object class recognition algorithms.
Evaluation of the latest classification algorithms based on convolutional neural network or deep networks.
Implementation of algorithms on cloud-enabled platforms.
Evaluation and implementation of the state-of-the-art RGBD recognition algorithms.
Development of fusion algorithms for a variety of sensors (cameras, IMU, odometry)
Development of robot navigation primitives for path planning, obstacle avoidance, and visual serving.
Development of SLAM systems using information from a variety of sensors (cameras, RGBD,IMU, odometry).
Prototype, interface, and integrate alternative sensors (e.g., inertial motion sensors, dead-reckoning sensors, RGBD cameras) into novel robot assemblies.
The typical work cycle involves research and algorithm development (typically in Python/Matlab), algorithm testing, refinement and benchmarking, followed by C/C++ code implementation and integration into the final application. Alternatively, the development cycle entitles the assembly of an electro-mechanical system, implementation of SW for integration of different components, and evaluation of the desired application.
3rd or 4th year BS student or graduate student.
Excellent mathematical background, with solid foundation in probability theory, systems theory, multivariate calculus, nonlinear optimization.
Proficient in C/C++ and Matlab (or Python), with good documentation and code organization habits.
Experience with pattern recognition and computer vision a definite plus. Experience with robotics and control systems a definite plus.
Creative and independent thinker, willing to take on challenges and learn new fields.
Work well in a fast-paced, small team environment.
If you’re looking for a job where you can exercise your mathematical and algorithmic skills, and are interested in doing work that can lead to graduate studies in computer vision and/or robotics, this is it!
We are the leading global consumer robot company, designing and building robots that empower people to do more, both inside and outside of the home. To date, we have sold over 25 million robots and globally employ more than 1,000 of the robot industry’s best and brightest. iRobot is committed to fostering invention, discovery and technological exploration in the pursuit of practical and valuable robot products for the home.