Autonomous underwater vehicles (AUVs) show much promise in environmental sensing, aquaculture, and security applications. Robust and adaptive control strategies can immensely benefit these scenarios by increasing autonomy and endurance. However, AUVs are nonlinear systems whose dynamics are challenging to model, especially during agile maneuvers at high angles of attack. To better capture these nonlinear effects, this paper proposes a physics-informed system identification scheme that combines prior knowledge of the system dynamics with data-driven regression. Strategies including Sparse Identification of Nonlinear Dynamics (SINDy), nonlinear least squares regression, and Gaussian processes (GPs) are used to learn the AUV dynamics online from measured data. These data-driven models are then implemented in an adaptive model predictive controller (MPC) for agile maneuvering that drives the system to a set point while updating the prediction model when new measurements are available. The performance of these three system identification strategies is evaluated on two different 6-DOF AUV platforms. All three strategies show good real-time performance, while the GP model offers the best balance between accuracy, speed and robustness. Field experimental data from the SAM AUV and the MOLA AUV are used for performance evaluation.

Hydrobatic autonomous underwater vehicles (AUVs) can be efficient in speed and range as well as agile in maneuvering, thereby enabling new use cases in ocean production, environmental sensing, and security. However, such robots are underactuated, have highly nonlinear dynamics at high angles of attack, and will be used in applications with high requirements for robustness. This paper explores the use of reinforcement learning (RL) to control hydrobatic AUVs, using the agile SAM AUV as a case study. The focus is on controlling the depth and pitch simultaneously, where there is a tight coupling between the states. This maneuver offers a simple, yet interesting test case to compare different control strategies. The twin-delay deep deterministic policy gradient (TD3) algorithm is applied to this AUV control problem. The resulting trained RL controller offers good robustness to noise and performs at a similar level as a Proportional-Integral-Derivative (PID) controller within the Stonefish simulation environment. The agent is also deployed and run on the robot hardware, with high overshoot. While the RL agent has good performance in simulation, the transfer from simulation to reality still leaves some open questions.

Agile and hydrobatic maneuvering capabilities can enhance AUV operations in increasingly challenging scenarios. In this paper, we explore the ability of an underactuated AUV to transition to and hold a pitch angle close to 90 degrees at a particular depth, like an inverted pendulum. Holding such an orientation can be valuable in observing a calving glacier, under-ice launch and recovery, underwater docking, inspecting vertical structures, and observing targets above the water surface. However, such control is challenging because of underactuation, rapid response times and varying stability in different configurations. To address this, a control policy is derived offline using nonlinear MPC in a high-fidelity simulation environment in Simulink. For real-time control, a hybrid controller using a behavior tree (BT) is developed based on the optimal MPC policy and applied on the AUV system. The BT controller considers Safety, Transit and Stabilize behaviors. The control algorithm is validated with simulations in Simulink and Stonefish-ROS as well as field experiments with the hydrobatic SAM AUV, showing repeatable performance in the inverted pendulum maneuver.

The term hydrobatics refers to the agile maneuvering of underwater vehicles. Underwater robots such as autonomous underwater vehicles (AUVs) and remotely operated vehicles (ROVs) are either designed as flight style, optimized for range and speed, or hover style, optimized for precise maneuverability. Hydrobatic capabilities can help balance efficiency and maneuverability in these platforms, enabling innovative robot designs for impact areas in environmental monitoring, ocean production and security. This dissertation addresses technical challenges related to hydrobatic AUVs and contributes to new knowledge in real-time control, simulation, learning and planning. 

Hydrobatic AUVs are underactuated systems --- new strategies using nonlinear model predictive control (MPC) and behavior trees (BTs) are presented for efficient and safe real-time control of underactuated AUVs in agile maneuvers. Further, the flow around an AUV during such maneuvers transitions from laminar to turbulent flow at high angles of attack, rendering flight dynamics modelling difficult. A full 0-360 degree envelope flight dynamics model is therefore derived, which combines a multi-fidelity hydrodynamic database with a generalized component-buildup approach. Such a model enables real-time (or near real-time) simulations of hydrobatic maneuvers including loops, helices and tight turns. To increase the intelligence and robustness of such systems, data driven methods including physics-informed learning, Gaussian processes, sparse regression  and reinforcement learning are utilized to rapidly identify models of the system's dynamics and perform online adaptive control. To further enhance autonomy, informative path planning is also studied, where an adaptive sampling strategy combines AUV measurements and satellite data to track ocean fronts.

These hydrobatic capabilities are safely brought to the real world through a cyber-physical system (CPS). Simulator environments are closely integrated with the robotic system, enabling pre-validation of controllers and software before hardware deployment. The small and hydrobatic SAM AUV (SAM: Small and Affordable Maritime robot) developed in-house at KTH as part of the Swedish Maritime Robotics Centre (SMaRC) is used as a test platform. The CPS concept is demonstrated with the SAM AUV in applications including detecting underwater targets, inspecting seaweed farm infrastructure and tracking algal blooms using the presented simulation, planning and control strategies.

Hydrobatik avser förmågan att utföra avancerade manövrar med undervattensfarkoster. Undervattensrobotar som autonoma undervattensfarkoster (AUV) är antingen optimerade för räckvidd och hastighet, eller optimerade för precisionsmanövrering. Hydrobatiska kapaciteter kan hjälpa till att balansera effektivitet och manövrerbarhet på dessa plattformar. Hydrobatik möjliggör innovativ robotdesign inom tre nyttoområden --- miljöövervakning, havsproduktion och säkerhet.I denna avhandling undersöks fördelar och tekniska utmaningar relaterade till hydrobatik. Avhandlingen bidrar till ny kunskap kring reglering, simulering, lärande och ruttplanering. Vidare tillämpas denna kunskap inom experiment av dessa robotar i realistiska scenarier.

Inom nämnda nyttoområden har ett antal scenarios identifierats där en kombination av manövrerbarhet samt räckvidd är avgörande för robotens förmåga att utföra sin uppgift. För att åstadkomma detta måste viktiga tekniska utmaningar lösas. För det första har dessa AUVer färre styrytor/trustrar än frihetsgrader, vilket leder till utmaning med underaktuering. Lösningsstrategier baserade på ickelinjär modelprediktiv kontroll (MPC) och beteendeträd (BTs) presenteras för effektiv och säker realtidskontroll av underaktuarande AUV:er i smidiga manövrar. För det andra är flödet runt en AUV som genomför hydrobatiska manövrar komplext. Övergången från laminärt till starkt turbulent flöde vid höga anfallsvinklar gör flygdynamikmodellering svår. En full 0-360 graders flygdynamikmodell härleds därför, vilken kombinerar en multi-tillförlitlighets hydrodynamisk databas med en generaliserad strategi för komponentvis-superpositionering av laster. Detta möjliggör prediktering av hydrobatiska manövrar som t.ex.  looping, roll, spiraler och väldigt snäva svängradier i realtids- eller nära realtids-simuleringar. För att öka intelligensen och robustheten hos sådana system används datadrivna metoder inklusive fysikinformerad inlärning, Gaussiska processer, sparsam regression och förstärkningsinlärning för att snabbt identifiera dynamiska modeller och utföra adaptiv kontroll i realtid. För att ytterligare förbättra autonomin studeras också informativ ruttplanering, där en adaptiv provtagningsstrategi kombinerar AUV-mätningar och satellitdata för att följa och mäta algblomningar och havsfrontar.

Dessa hydrobatiska förmågor överförs på ett säkert sätt till den verkliga världen genom ett cyberfysiskt system (CPS). Simulatormiljöer är integrerade med robotsystemet, vilket möjliggör förvalidering av styrenheter och mjukvara innan hårdvaruinstallation. Den lilla och hydrobatiska AUV:n SAM (SAM: Small and Affordable Maritime robot), egenutvecklad på KTH som en del av Swedish Maritime Robotics Centre, används som testplattform. CPS-konceptet demonstreras under fältförhållanden med SAM AUV. Applikationer inkluderar sökuppdrag av ett nedsänkt föremål, inspektioner av infrastruktur för havsbruk samt spårning av algblomning.

Hydrobatic autonomous underwater vehicles (AUVs) can be efficient in range and speed, as well as agile in maneuvering. They can be beneficial in scenarios such as obstacle avoidance, inspections, docking, and under-ice operations. However, such AUVs are underactuated systems—this means exploiting the system dynamics is key to achieving elegant hydrobatic maneuvers with minimum controls. This paper explores the use of model predictive control (MPC) techniques to control underactuated AUVs in hydrobatic maneuvers and presents new simulation and experimental results with the small and hydrobatic SAM AUV. Simulations are performed using nonlinear model predictive control (NMPC) on the full AUV system to provide optimal control policies for several hydrobatic maneuvers in Matlab/Simulink. For implementation on AUV hardware in robot operating system, a linear time varying MPC (LTV-MPC) is derived from the nonlinear model to enable real-time control. In simulations, NMPC and LTV-MPC shows promising results to offer much more efficient control strategies than what can be obtained with PID and linear quadratic regulator based controllers in terms of rise-time, overshoot, steady-state error, and robustness. The LTV-MPC shows satisfactory real-time performance in experimental validation. The paper further also demonstrates experimentally that LTV-MPC can be run real-time on the AUV in performing hydrobatic maneouvers.

This paper outlines challenges and opportunities in operating underwater robots (so-called AUVs) on a seaweed farm. The need is driven by an emerging aquaculture industry on the Swedish west coast where large-scale seaweed farms are being developed. In this paper, the operational challenges are described and key technologies in using autonomous systems as a core part of the operation are developed and demonstrated. The paper presents a system and methods for operating an AUV in the seaweed farm, including initial localization of the farm based on a prior estimate and dead-reckoning navigation, and the subsequent scanning of the entire farm. Critical data from sidescan sonars for algorithm development are collected from real environments at a test site in the ocean, and the results are demonstrated in a simulated seaweed farm setup.

This study combines high-fidelity simulation models with experimental power consumption data to evaluate the performance of Energy Management Strategies (EMS) for fuel cell/battery hybrid Autonomous Underwater Vehicles (AUV). The performance criteria are energy efficiency, power reliability and system degradation. The lack of standardized drive cycles is met by the cost-efficient solution of synthesizing power profiles from sampled AUV field trial data. Three power profiles are used to evaluate finite-state machine, fuzzy logic and two optimization-based EMS. The results reveal that there is a trade-off between the objectives. The rigidity of the EMS determines its load-following behavior and consequently the performance regarding the objectives. Rule-based methods are particularly suitable to design energy-efficient operations, whereas optimization-based methods can easily be tuned to provide power reliability through load-following behavior. Both classes of EMS can be best-choice methods for different types of missions.

Hydrobatic AUVsare very agile, and can perform challenging maneuvers that encompass the full 0 degrees-360 degrees flight envelope. Such AUVs can be beneficial in novel use cases in ocean production, environmental sensing, and security, by enabling new capabilities for docking, inspection, or under-ice operations. To further explore their capabilities in such scenarios, it is crucial to be able to model their flight dynamics over the full envelope, which includes strong nonlinear effects and turbulence at high angles of attack. With accurate and efficient simulation models, new hydrobatic maneuvers can be generated and control strategies can be developed. Therefore, this article contributes with a strategy to perform efficient and accurate simulations of hydrobatic maneuvers in real time. A multifidelity hydrodynamic database is synthesized by combining analytical, semiempirical, and numerical methods, thereby capturing fluid forces and moments over the full envelope. A component buildup workflow is used to assemble a nonlinear flight dynamics model using lookup tables generated from the database. This simulation model is used to perform real-time simulations of advanced hydrobatic maneuvers. Simulation results show agreement with literature and experiment, and the simulator shows utility as a development tool in designing new maneuvers and control strategies.

Cyber-physical systems (CPSs) comprise a network of sensors and actuators that are integrated with a computing and communication core. Hydrobatic Autonomous Underwater Vehicles (AUVs) can be efficient and agile, offering new use cases in ocean production, environmental sensing and security. In this paper, a CPS concept for hydrobatic AUVs is validated in real-world field trials with the hydrobatic AUV SAM developed at the Swedish Maritime Robotics Center (SMaRC). We present system integration of hardware systems, software subsystems for mission planning using Neptus, mission execution using behavior trees, flight and trim control, navigation and dead reckoning. Together with the software systems, we show simulation environments in Simulink and Stonefish for virtual validation of the entire CPS. Extensive field validation of the different components of the CPS has been performed. Results of a field demonstration scenario involving the search and inspection of a submerged Mini Cooper using payload cameras on SAM in the Baltic Sea are presented. The full system including the mission planning interface, behavior tree, controllers, dead-reckoning and object detection algorithm is validated. The submerged target is successfully detected both in simulation and reality, and simulation tools show tight integration with target hardware.

This paper presents a comparison of different energy management strategies (EMS) for autonomous underwater vehicles (AUV) with fuel cell-battery hybrid power systems. Sophisticated EMS can decrease energy consumption, limit fuel cell degradation or increase reliability. EMS for hybrid vehicles have been studied extensively in the automotive industry where standardised drive cycles are applied. As for AUVs, there are no standard drive cycles and power profiles can vary significantly depending on the type of mission. In this study, rule-based and optimization-based EMS are compared. The rule-based strategies rely on deterministic rules and fuzzy logic, the optimization-based strategies minimize a constrained cost function to efficiently split the power demand. The EMS are evaluated against a previously sampled power profile of a Hugin 3000 AUV. The evaluation against real power profiles adds significant value to this study.