Seeing Around Corners Using Smartphone-Grade Lidar

2026-05-20 · Source: IEEE Spectrum · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Advanced, short

Summary

A new study demonstrates that off-the-shelf smartphone-grade lidar, costing less than US \$100, can effectively detect objects hidden around corners, a capability previously limited to expensive lab-grade devices costing US \$0.5 million to \$1 million. This advance holds potential for autonomous driving, enhancing safety at blind intersections by detecting vehicles, cyclists, or pedestrians, and for robotics, aiding navigation in cluttered environments. Researchers achieved this by analyzing multiple lidar images simultaneously, drawing inspiration from burst photography and synthetic aperture radar, and developing algorithms to combine faint signals. The portable smartphone lidar system, featuring about 100 pixels, successfully reconstructed 3D images of static hidden objects, tracked 3D motions, and pinpointed the sensor's location. While currently recovering sparse geometric and motion data rather than detailed photographic images, the team has publicly released the necessary code, with findings detailed online 20 May in Nature.

Key takeaway

For Robotics Engineers or Autonomous Driving System Developers evaluating perception sensor suites, this research indicates that low-cost, smartphone-grade lidar is now a viable option for non-line-of-sight sensing. You can enhance your system's situational awareness by detecting hidden obstacles around corners, improving safety and navigation in complex environments. Explore the publicly released code to assess its integration potential for your specific application needs.

Key insights

Smartphone-grade lidar, enhanced by algorithms, can perform non-line-of-sight imaging, making advanced sensing accessible.

Principles

• Consumer lidar captures faint around-the-corner signals.
• Combining multiple measurements clarifies noisy data.
• Accessible tech drives diverse, unforeseen applications.

Method

The method analyzes multiple noisy lidar images, combining information across measurements using algorithms inspired by burst photography and synthetic aperture radar to reveal hidden signals.

In practice

• Enhance autonomous vehicle safety at blind spots.
• Aid robot navigation in cluttered environments.
• Orient sensors using hidden objects as landmarks.

Topics

• Lidar
• Non-Line-of-Sight Imaging
• Smartphone Lidar
• Autonomous Driving
• Robotics
• Signal Processing

Code references

• sidsoma/consumer-nlos

Best for: Computer Vision Engineer, AI Scientist, Robotics Engineer, Research Scientist

Related on AIssential

• SurroundNEXO: Ego-Centric Metric Bridging for Spatially Consistent Geometry in Autonomous Driving — SurroundNEXO bridges low-overlap multi-camera views for metric depth prediction using ego-centric geometry and sparse LiDAR anchoring.
• RISE: Single Static Radar-based Indoor Scene Understanding — RISE utilizes multipath radar reflections and generative AI for privacy-preserving, geometry-aware indoor scene understanding from a single static sensor.
• Taming Perception Jitter: Uncertainty-Aware LiDAR Object Detection for Reliable Motion Classification — Augmenting LiDAR object detection with uncertainty estimates and statistical tests effectively distinguishes true motion from perception jitter.
• Why Rivian Uses BOTH Cameras and LiDAR — Multiple, diverse sensors provide robust, complementary environmental perception, overcoming individual sensor limitations.
• Heterogeneous LiDAR Early Fusion and Learned Re-Ranking Strategy for Robust Long-Term Place Recognition in Unstructured Environments — Heterogeneous LiDAR early fusion and learned re-ranking significantly boost place recognition in challenging unstructured agricultural environments.
• Vision-Based Human Awareness Estimation for Enhanced Safety and Efficiency of AMRs in Industrial Warehouses — Estimating human awareness via vision allows AMRs to adapt behavior, improving safety and efficiency in shared spaces.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by IEEE Spectrum.