Active Safety Envelopes using Light Curtains with Probabilistic Guarantees

• DOI: 10.15607/rss.2021.xvii.045
• Publication Date: 2021-06-27T13:30:56Z
• Authors: Siddharth Ancha; Gaurav Pathak; Srinivasa Narasimhan; David Held
• Keywords: Envelope (radar)

To safely navigate unknown environments, robots must accurately perceive dynamic obstacles.Instead of directly measuring the scene depth with a LiDAR sensor, we explore use much cheaper and higher resolution sensor: programmable light curtains.Light curtains are controllable sensors that sense only along surface user selects.We to estimate safety envelope scene: hypothetical separates robot from all obstacles.We show generating random locations (from particular distribution) can quickly discover for scenes objects.Importantly, produce theoretical guarantees on probability detecting an obstacle using curtains.We combine machine learning based model forecasts tracks motion efficiently.Our method estimates envelopes while providing probabilistic be used certify efficacy perception system detect avoid evaluate our approach in simulated urban driving environment real-world moving pedestrians curtain device efficiently effectively. 1