Products
Autonomy software
Skarv Technologies proprietary autonomy architecture, navigation solution, and unmatched situational awareness (driven by real-time data streams) provide an ideal platform to automate a wide range of tasks underwater, such as autonomous net- and hull cleaning, coverage mapping, pipe tracking, subsea inspection, etc. We also provide adaptive mission planning capabilities to support long duration missions at full ocean depth, as well as simulation and optimization software. Reduce time and cost while also increasing the prospect of retrieving relevant data. Our software is flexible and can be implemented on various platform and tailored to perform a wide range of tasks.
Features
Mission planning and execution
Plan, automate, and execute missions.
Situational awareness
Combine data streams to provide situational understanding.
Obstacle avoidance
Dynamic obstacles generation and re-planning.
Module based
Software modules can be introduced into existing systems.
Human in the loop
Control and monitor automation and mission execution.
Risk aware decision making
Ensure the safety of operation and platform integrity.
Platform independent
Can be integrated on multiple platforms.
Services
Data analysis tools
We offer consultancy services and tools for doing complex data processing and analysis. This includes modern machine vision and intelligence methods such as: machine learning, object detection, segmentation, and classification. We specialize on data streams captured in the ocean such as 2D/3D video, point clouds, hyperspectral imaging, acoustic imaging and communications, and typical time-series data.
Features
Machine vision
Detect, segment, and track objects on the seafloor or on structures.
Orthomosaics
Images processing to georeferenced and orthorectified maps.
Underwater image enhancement
De-hazing and removal of backscatter.
3D Models (photogrammetry)
Structure from motion methods for 3D reconstruction of underwater scenes.
Underwater hyperspectral imaging
Processing of hyperspectral images.
Processing for aquaculture
Autonomous net following and hole detection.
Acoustic processing
Processing of multibeam and side scan sonar data.
Processing for offshore surveys
Mission optimization and online data interpretation.
Time-series
Analysis and processing of time series data.
Processing for environmental surveys
Tools for benthic habitat mapping and monitoring
Autonomous platform services and survey support
We offer help with customized sensor- and software integrations for marine robotic solutions. Based on over 15 years experience doing autonomous underwater operations at various depths and in the Arctic, using USVs, AUVs, and ROVs. We can help turning concepts into solutions and offer expertise and support for complex survey missions where autonomy is required.
Data visualization
Do you have large data structures that you either need to visualize or work with interactively? We can provide web-based portals to explore, query, visualize, and analyse geospatial data. Gain overview and new insight into your data and share this with clients or public users. We have experience with spatio-temporal data formats such as netCDF and HDF5, GIS data, as well as conventional video and imaging data types.
Examples from use cases
Autonomous mapping
Using the Skarv Autonomy Engine (SKATE) we are able to autonomously map the seafloor and allow for adaptive optimization schemes that can increase the amount and relevance of the data being collected. The system automatically adjusts the speed and height to ensure that the quality of the data is maintained. Using Skarvs own stereo-camera system we are able to generate photo-mosaics with resolutions below 0.5cm (shown in the image to the left).
Automatic detection of known objects
Using custom trained artificial neural nets we are able to detect, segment, and track objects on the seafloor or on structures. The systems can be trained on new data to do custom detection of underwater objects such as debris, pipelines, anchors, archaeology, damage types, etc.
Anomaly detection of new objects
Finding unknown object without prior information is also possible using attention-based neural architectures. The system can be used to look for lost objects, damages, irregularities, and conspicuous elements.
3D reconstruction of underwater scenes
Using structure from motion (SfM) and CNN-based feature descriptors we are able to reconstruct 3D-information from difficult underwater scenes amidst schools of fish and suspended particles. Our tools can deliver geo-referenced 3D models in standard formats for integration into BIM-models.