Authors: Devansh R Agrawal, Rajiv Govindjee, Jiangbo Yu, Anurekha Ravikumar, Dimitra Panagou
Published on: February 07, 2024
Impact Score: 8.45
Arxiv code: Arxiv:2402.05254
Summary
- What is new: Two novel algorithms for certified perception in robotics, focusing on safety-critical applications with results demonstrated at 30FPS.
- Why this is important: The lack of certified visual odometry and mapping algorithms that can accurately and safely navigate robots in environments with potential hazards.
- What the research proposes: Introducing a certified visual odometry algorithm ensuring provable error bounds and a certified mapping algorithm safely underestimating distances to obstacles.
- Results: Successful demonstration in hardware experiments, showing superior performance and safety compared to state-of-the-art techniques.
Technical Details
Technological frameworks used: RGBD camera-based perception
Models used: Certified Visual Odometry, Certified Mapping algorithm
Data used: RGBD images
Potential Impact
Robotics, autonomous vehicle companies, and safety-critical systems industries.
Want to implement this idea in a business?
We have generated a startup concept here: SafePath Robotics.
Leave a Reply