Anschütz Achieves Breakthrough with New COLREG Navigation Algorithm
Anschütz, a prominent global authority in navigation and bridge systems, has effectively executed and evaluated an automatic Collision Avoidance System (CAS) as part of the Kiel CAPTN initiative. This achievement signifies another significant advancement toward autonomous navigation and assistance technologies for safer maritime operations.
The CAS employs an algorithm that scrutinizes the maritime landscape utilizing traditional onboard sensor data to detect possible collision threats with other vessels. In accordance with the Convention on the International Regulations for Preventing Collisions at Sea (COLREG), these threats are persistently evaluated and dynamically illustrated. When a collision threat is identified, the system instantaneously computes safe maneuvering zones and proposes routes, factoring in water depths, navigational aids, restricted zones, and other pertinent elements.
“The real-world evaluations in the distinctive test environment of the Kiel CAPTN initiative have provided us with the chance to merge our research laboratory and simulator insights with practical experiences and user feedback over a span exceeding nine months,” stated Daniel Sommerstedt, Project Manager for autonomous navigation research at Anschütz. “We have consistently acquired fresh insights, which we have been able to directly incorporate into enhancing the algorithm and streamlining the display.”
The fully rule-based COLREG algorithm demonstrated remarkable reliability, efficiency, and fine-tuning during evaluations across varying conditions and scenarios. Target identification and allocation, situational assessment, visualization of potential collision threats, trajectory computation, acknowledgment, and incorporation of the motion control system combining trackpilot and speedpilot functioned seamlessly together.
The system’s situational evaluations and recommended actions corresponded with those of the navigators present. Trials on the Kiel Fjord unveiled challenging scenarios that the CAS successfully tackled with dependable recommendations for actions, enhancing its performance progressively.
“Automating target detection, correlation, and evaluation in a constrained and bustling area like the Kiel Fjord proved to be a challenge,” Sommerstedt commented. “We have made significant strides along a steep learning curve and will continue to refine and evolve the process as part of the recently launched X-Ferry project within the CAPTN initiative.”
In this upcoming phase, the situational analysis will be further enriched with camera data. In addition to detecting potential supplementary objects, intelligent cameras will categorize existing entities, thereby verifying the radar and AIS evaluations and providing a more comprehensive maritime overview.
Anschütz aims to create a market-ready, certified assistance system for professional shipping. Over the next year, Anschütz will persist in testing and refining the CAS as part of the X-Ferry research project. The system will also be adapted for various prototypes, test vessels, and simulators to accumulate a broad range of test scenarios, advancing the CAS to full maturity for certification.
As part of onboard assistance systems, the CAS will enhance situational awareness and alert navigators to potential collision threats. This advancement will alleviate workload during high-stakes situations and minimize human error. In certain cases, it may also facilitate a reduced bridge crew, addressing the shortage of qualified personnel.
Anschütz remains dedicated to propelling autonomous navigation through various research initiatives, including the CAPTN initiative in Kiel and diverse customer projects.
The COLREG algorithm was developed within the OCUMAR research initiative in collaboration with project partner KIT (Karlsruhe Institute of Technology) and was financed by the German Federal Ministry for Economic Affairs and Climate Action.
