Urgent.News

What's breaking now, across thousands of outlets.

Tech

Forward Watch

AI Object Detection for Recreational Boats Every skipper knows the moment: a log just under the surface, a kayak hidden in sun glare, or a small boat with no AIS crossing ahead at dusk. Keeping a proper lookout is a legal and practical duty, but human eyes get tired. Forward Watch is an AI-powered assistant to that lookout. It is an open-source Signal K plugin , published on npm as…

Forward Watch is an AI-powered safety assistant for recreational boat skippers. It continuously monitors a camera mounted on the bow of the vessel, using object detection to identify hazards that electronic sensors might miss. Forward Watch is an open-source Signal K plugin published on npm, and part of the d3kOS platform. The plugin detects six classes of objects: ships, boats, recreational crafts, floating debris, buoys, and kayaks.

It is designed to be run on a Raspberry Pi 4 or better, with no need for a GPU, and works by analyzing frames captured from the camera at a configurable interval. Forewarning only raises an alert when an object is repeatedly detected, or when it is close to the boat, to avoid false alarms. The plugin provides various ways for the skipper to be alerted to potential hazards, including Signal K data, notifications, chart targets, and support for standalone chart plotters.

Forward Watch is designed as an aid to the human lookout, not a replacement, and has been tested with positive results in real-world conditions.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in Tech

Surviving AWS Spot Interruptions With Real Checkpointing

🔧 The Problem Spot instances are 60-90% cheaper than on-demand, and the pitch is always the same: "just make your workload fault-tolerant." Great, thanks.

  • Implement checkpointing pattern to mitigate spot interruptions
  • Checkpoint writer serializes progress to durable storage
  • Interruption watcher triggers emergency checkpoint on termination notice

Same model, same defences: 7 document attacks down to 0, chat attacks barely moved

A prompt-injection benchmark, three bugs in my own scorer, and the result I only found after fixing them. I built RAGProbe, an open-source tool that throws prompt-injection attacks at LLM apps and…

  • Seven document attacks successfully blocked by layered defenses
  • Chat attacks barely increased from 37 to 40 after defenses
  • Three bugs and one incorrect claim in original attack tool fixed

More from Monday 2 November →