Sensor Fusion Engineer Intern at Endiatx

Overview

Duration: 3-month internship with potential for full-time conversion

Endiatx is a pioneering medical device company based in Hayward, California, developing cutting-edge microrobotics for diagnostic applications in the human body. Our flagship product, Pill Bot™, is a propeller-driven endoscopic capsule designed to transform gastrointestinal screening. We are a multidisciplinary team of innovators, engineers, clinicians, and designers working together to change the future of medical diagnostics.

Endiatx is seeking a detail-oriented and innovative Sensor Fusion Engineer Intern to join our team for the summer of 2025. This internship is ideal for an undergraduate or graduate student with a strong foundation in robotics, computer vision, estimation theory, and signal processing. You will be responsible for developing algorithms to combine data from the Pill Bot™'s onboard Inertial Measurement Unit (IMU) and front-facing camera to produce a robust, drift-free estimate of the robot's position, velocity, and orientation within the stomach environment.

• Develop sensor fusion algorithms: Design and implement algorithms to combine IMU data (acceleration, angular velocity) with visual information from the capsule's camera.
• Implement state estimation techniques: Apply and adapt techniques such as Kalman filters (e.g., EKF, UKF) to estimate the robot's position and orientation.
• Incorporate visual data for drift correction: Utilize camera images for visual odometry or other methods to correct for IMU drift and enhance localization accuracy.
• Evaluate performance: Test and validate the sensor fusion system's accuracy and robustness using simulated and potentially real-world data.
• Collaborate with ML/CV team: Work with the Machine Learning/Computer Vision intern to ensure complementary approaches to robot localization and environmental understanding.

• Pursuing or completed a degree in Robotics, Electrical Engineering, Computer Science, or a related field.
• Strong academic background in estimation theory, signal processing, and control systems
  .
• Experience with Kalman filters (EKF, UKF) or similar probabilistic state estimation techniques.
• Familiarity with computer vision concepts and processing camera data for pose estimation (e.g., visual odometry).
• Proficiency in programming languages like Python or C++.
• Analytical and problem-solving skills, with a focus on practical application in a robotics context.