CIR: Robot Arm Collision Avoidance and Onboard Sensors
Sensors are required to detect obstacles near a robot manipulator. While depth cameras are the most common scene perception technology, we look at alternatives that give information originating from the robot body itself (on-board sensors). For robots to be physically located within human–laden environments, advances are needed in both control and sensing technology. We will look at formulations and execution of real-time control methods for robot manipulator collision avoidance around dynamic objects and how to perceive objects in real-time.
Date and Time
Location
Hosts
Registration
- Date: 15 Nov 2023
- Time: 06:00 PM to 07:00 PM
- All times are (UTC-07:00) Mountain Time (US & Canada)
-
Add Event to Calendar
- 2155 E Wesley Ave
- Denver, Colorado
- United States 80208
- Building: University of Denver Ritchie School of Engineering and Computer Science
- Room Number: 410
- Click here for Map
- Starts 23 October 2023 04:14 PM
- Ends 15 November 2023 08:14 PM
- All times are (UTC-07:00) Mountain Time (US & Canada)
- No Admission Charge
Speakers
Caleb Escobedo
Robot Arm Collision Avoidance and Onboard Sensors
Sensors are required to detect obstacles near a robot manipulator. While depth cameras are the most common scene perception technology, we look at alternatives that give information originating from the robot body itself (on-board sensors). For robots to be physically located within human–laden environments, advances are needed in both control and sensing technology. We will look at formulations and execution of real-time control methods for robot manipulator collision avoidance around dynamic objects and how to perceive objects in real-time.
Biography:
Caleb Escobedo is a fifth year Computer Science PhD student at the University of Colorado Boulder. Mr. Escobedo’s research focuses on novel close-proximity (< 15cm) sensor development and sensor integration with dynamic robot arm movement. Mr. Escobedo spent the last year working at the Samsung Artificial Intelligence Center – New York on the Novel Sensors Team, advised by Daewon Lee and Volkan Isler. Within Samsung, Mr. Escobedo published three papers about using piezoelectric–based sensors to detect close–proximity and contact with nearby objects at the International Conference on Robotics and Automation (ICRA) and the International Conference on Intelligent Robots and Systems (IROS) 2023 and is preparing for additional submissions at ICRA 2024.
Mr. Escobedo is member of the Human Interaction and Robotics (HIRO) group at the University of Colorado Boulder, headed by Alessandro Roncone, where he leads the HIRO Robotic Skin Project in development of software and hardware required for full-surface robot close–proximity and contact detection. Mr. Escobedo has developed several software stacks integrating real–time redundant robot manipulator inverse kinematics with data from custom sensors placed on the robot surface, and his recent publications with the HIRO group include a IROS 2023 acceptance on trajectory planning with contact in cluttered environments, and a submission to ICRA 2024 introducing custom hardware for tactile sensing. While in New York, NY. Mr. Escobedo was the lead student organizer for the Robotics: Science and Systems (RSS) 2022 workshop titled Close–Proximity Human–Robot Collaboration: Challenges and Opportunities hosted at Columbia University.