Browse Topic: Adaptive control
Deep Reinforcement Learning (DRL) for quadrotor flight control typically relies on Domain Randomization (DR) for sim-to-real transfer, resulting in overly conservative policies that struggle with dynamic disturbances. To overcome this, we propose a novel adaptive control architecture that actively perceives and reacts to instantaneous perturbations. First, we train an optimal outer-loop policy, then replace its reliance on ground-truth disturbance data with a Residual Dynamics Predictor (RDP). The RDP estimates the external forces and moments acting on the aircraft in flight online using only the history of states and control actions. For seamless hardware transfer, we introduce a data-efficient linear calibration bridge and an online thrust correction mechanism that align the simulated latent space with reality using mere seconds of flight data. Real-world validations on a Crazyflie micro-quadrotor demonstrate that our adaptive controller significantly outperforms baselines, maintaining precise trajectory tracking under severe uncertainties including mass variations, asymmetric payloads, and dynamic slung loads.
Accurate control of the engine park angle during Autostop in hybrid vehicles is critical for enabling rapid and smooth Autostarts, reducing start-up vibrations, and enhancing overall driving comfort. However, in real-world scenarios, the available torque for engine positioning is often limited by competing driver torque demands, battery discharge constraints, and the state of charge (SoC). Under these conditions, conventional position-speed control strategies frequently fail to achieve the desired precision. This paper introduces an adaptive control strategy for the electric machine (EM) that drives the internal combustion engine, ensuring precise alignment of the crankshaft at a predefined angle to optimize restart conditions. Upon receiving an engine shutdown request, the proposed controller computes an adaptive deceleration profile that respects the EM’s torque and deceleration limits while guiding the crankshaft toward the target park position. The core of the approach lies in generating a theoretical speed trajectory and tracking it through an adaptive nonlinear control law that dynamically adjusts in real time to compensate for disturbances and eliminate residual angular error at the end of the maneuver. Unlike conventional methods, the proposed solution maintains robustness under stringent deceleration constraints and varying operating conditions. Simulation and experimental results on a hybrid powertrain test bench demonstrate that the proposed method significantly improves park angle accuracy and consistency even under limited deceleration scenarios.
Electric Vehicles (EVs) are rapidly transforming the automotive landscape, offering a cleaner and more sustainable alternative to internal combustion engine vehicles. As EV adoption grows, optimizing energy consumption becomes critical to enhancing vehicle efficiency and extending driving range. One of the most significant auxiliary loads in EVs is the climate control system, commonly referred to as HVAC (Heating, Ventilation, and Air Conditioning). HVAC systems can consume a substantial portion of the battery's energy—especially under extreme weather conditions—leading to a noticeable reduction in vehicle range. This energy demand poses a challenge for EV manufacturers and users alike, as range anxiety remains a key barrier to widespread EV acceptance. Consequently, developing intelligent climate control strategies is essential to minimize HVAC power consumption without compromising passenger comfort. These strategies may include predictive thermal management, cabin pre-conditioning, zonal climate control, and integration with renewable energy sources. By implementing such energy-efficient solutions, EVs can achieve better range performance, improved user satisfaction, and greater environmental benefits. Modern EV climate control systems increasingly rely on intelligent features such as Auto mode, which dynamically adjusts fan speed, airflow direction, and temperature settings based on real-time cabin and ambient conditions. By leveraging sensor data and adaptive control algorithms, Auto mode optimizes thermal comfort while minimizing unnecessary energy expenditure. This automated regulation plays a crucial role in reducing HVAC-related power consumption, thereby contributing to overall range improvement and enhancing system efficiency without compromising passenger comfort. This study focuses on the development of three distinct Auto mode calibration levels for each set condition, designed to achieve the same cabin temperature with varying dynamic responses and energy consumption profiles. In Auto mode, the cabin temperature is regulated through intelligent control of compressor speed, blower speed, and evaporator temperature. While all Auto levels can maintain the desired setpoint, the time required to reach this temperature and the system’s responsiveness to sudden thermal loads can vary significantly. This study introduces three distinct calibration profiles, each engineered to achieve the same cabin temperature under different dynamic conditions and energy consumption levels. These profiles allow users to choose between faster thermal response or reduced power usage, effectively enabling a trade-off between immediate comfort and extended driving range
Vertical Take-Off and Landing (VTOL) aircraft introduce complex monitoring challenges due to distributed propulsion, lightweight structures, and variable operating conditions. This paper presents advanced Frequency and Orders domain techniques that repurpose existing flight control, propulsion, and structural sensor data to enhance observability without additional instrumentation. By transforming vibration, acoustic, and electrical signals into frequency and order domains, the approach enables detection of harmonics, resonance, and fault signatures tied to rotor dynamics, supporting adaptive control and predictive maintenance. Beyond rotor systems, these techniques are equally effective for monitoring electric motor health, gearbox wear, bearing degradation, and structural coupling effects in composite airframes. They also provide insight into power electronics and thermal management systems by identifying spectral anomalies linked to electrical imbalance or cooling inefficiencies. Aggregated fleet data strengthens prognostic capabilities, enabling early detection of systemic issues and trend analysis. Applications include mitigating ground resonance and modal instabilities, as well as improving reliability of propulsion and structural subsystems. Integration into avionics emphasizes computational efficiency, scalability, and compliance with standards such as DO-160 [1], DO-178 [2], ARP4761 [3] and ARP4764 [4]. Simulation and bench testing confirm feasibility, demonstrating potential to enhance safety, reliability, and lifecycle cost for next-generation urban air mobility platforms.
In the context of increasing global energy demand and growing concerns about climate change, the integration of renewable energy sources with advanced modelling technologies has become essential for achieving sustainable and efficient energy systems. Solar energy, despite its considerable potential, continues to face challenges related to performance variability, limited real-time insights, and the need for reactive maintenance. To overcome these barriers, this work presents a Digital Twin framework aimed at optimizing solar-integrated energy systems through real-time monitoring, predictive analytics, and adaptive control. This work presents a Digital Twin framework designed to address the challenges of designing, operating, maintaining, and estimating renewable energy systems, specifically solar power, based on dynamic load demand. The framework enables real-time forecasting and prediction of energy outputs, ensuring systems operate efficiently and maintain peak performance across diverse conditions. The proposed methodology mirrors the physical system using real-time data inputs, environmental conditions, and physics-based models to create a high-fidelity virtual replica. This allows for dynamic analysis of energy flows, load forecasting, system performance prediction, and scenario testing to optimize design and operational strategies. By integrating predictive analytics, Digital Twin adapts to changing conditions, enabling proactive maintenance, fault detection, and system calibration to meet future load demands. Experimental validation demonstrates that the framework improves system efficiency, adaptability, and reliability, with scalable applications for both centralized and decentralized energy systems. Additionally, its integration with cloud-based platforms and IoT technologies enables real-time monitoring, facilitating continuous optimization and data-driven decision-making. This Digital Twin approach provides an intelligent, data-driven solution for the renewable energy sector, facilitating sustainable, resilient, and efficient energy infrastructures that can reliably meet evolving load demands while optimizing performance throughout their lifecycle.
In the power industry, high-power Diesel Generator (DG) sets often utilize high power V-engine cylinder configurations to enhance power output within a compact design, ensuring smoother operation and reduced vibration. In this V-engine configurations, the exhaust gas mass flow rate is significantly higher compared to inline engines of similar displacement, due to the greater number of cylinders operating in a compact space, which leads to a higher volume of exhaust gases expelled in a shorter duration. This necessitates the use of a dual Exhaust After Treatment System (EATS) to effectively manage NOx emissions. High-power gensets typically emit NOx levels around 9 g/kWh, presenting significant challenges for developers in adhering to stringent emission standards. To address these challenges and meet CPCB IV+ emission norms, we propose a dual urea dosing system integrated with a novel control strategy aimed at optimizing the treatment of exhaust gases. This paper introduces a dual exhaust system equipped with dual urea dosing units. By employing two controller units, we ensure compliance with On-Board Diagnostics (OBD) requirements while effectively implementing advanced software concepts. Our approach not only enhances the efficiency of NOx reduction but also provides a robust solution for high-power diesel generators, paving the way for more sustainable operations in the power sector. Furthermore, we explore the integration of real-time monitoring and adaptive control mechanisms to respond dynamically to varying load conditions and exhaust characteristics. This ensures optimal dosing of urea, enhancing the overall performance of the EATS. This study discusses the design, implementation, and performance evaluation of the proposed system, highlighting its potential to significantly lower NOx emissions while maintaining operational efficiency in high-power diesel generator applications.
Generalized Predictive Control (GPC) is an advanced form of an adaptive control algorithm that uses experimentally acquired data to determine the input-output relationship of complex systems through a process called system identification. GPC has historically been employed for stability augmentation and vibration reduction of dynamically-scaled tiltrotor aircraft wind-tunnel models since the complex nature of these dynamic systems does not lend itself well to traditional control approaches. The present research expands upon previous analytical and experimental work with wind-tunnel experiments that utilize improved GPC techniques. These techniques improved controller robustness such that a working controller was stable across a multitude of model configurations and wind-tunnel conditions and successfully suppressed vibration and vehicle flutter. Advanced GPC (AGPC) enables self-adaptation of a traditional GPC control law. AGPC was also investigated during the present research but was not needed as anticipated because of the robustness resulting from improvements made to traditional GPC.
The development of an adaptive pilot model for rotorcraft tracking tasks is useful to understand and replicate human pilot behavior under varying vehicle dynamics and environmental conditions. This paper presents a Model-Reference Adaptive Control (MRAC)-based pilot model designed to emulate the adaptability of human pilots during attitude and position tracking tasks. The model leverages wavelet analysis to characterize pilot behavior and employs a closed-loop system identification approach to derive baseline pilot parameters. MRAC methodology using state-feedback is implemented and validated through simulations involving time-varying vehicle dynamics, such as changes in control sensitivity and added phase delays. Results demonstrate the model's ability to maintain consistent tracking performance despite dynamic modifications, though discrepancies with human pilot data highlight the complexity of fully capturing adaptive human control strategies. The proposed model offers a framework for integrating human adaptability into flight system design and simulation tools.
Unmanned Underwater Vehicles (UUVs) are used around the world to conduct difficult environmental, remote, oceanic, defense and rescue missions in often unpredictable and harsh conditions. A new study led by Flinders University and French researchers has now used a novel bio-inspired computing artificial intelligence solution to improve the potential of UUVs and other adaptive control systems to operate more reliability in rough seas and other unpredictable conditions.
One of the challenges of Electric Vehicles (EVs) is to provide thermal comfort for the occupants while minimizing the energy consumption and the impact on the driving range. Conventional heating systems, such as Positive Temperature Coefficient (PTC) heaters, consume a large amount of battery power and reduce the efficiency of the EVs. Heat Pumps (HPs) are an alternative heating system that can divert heat from the ambient air and transfer it to the cabin. HPs can achieve higher Coefficient of Performance (COP) than PTC heaters and save energy. However, for Indian sub-continent conditions HPs have some drawbacks, such as low heating capacity at low ambient temperatures, and variable performance depending on the operating conditions. Therefore, it is important to design and control the HP system optimally. This study employs 1D Computer-Aided Engineering (CAE) modelling and simulation techniques to analyse the performance of heat pump systems within the confined environment of an EV cabin. The simulations explore various operating conditions and personalized comfort profiles to optimize the control strategy of the heat pump. By adapting to individual preferences and external environmental factors, the heat pump control system aims to enhance occupant comfort while minimizing energy consumption. The results indicate that the integration of adaptive thermal comfort principles in heat pump control systems enables significant improvements in both comfort levels and energy efficiency compared to conventional PTC heaters. The personalized comfort profiles allow for tailored temperature regulation, ensuring occupants experience optimal thermal conditions throughout their journey. Additionally, the adaptive control strategy optimizes the utilization of the heat pump, reducing energy wastage and extending the driving range of the EV. In conclusion, the adoption of heat pump technology with personalized comfort profiles presents a promising approach to enhance the thermal management capabilities of electric vehicles. This study contributes to the ongoing efforts in advancing sustainable transportation by prioritizing occupant comfort and energy efficiency simultaneously.
This research aims to develop an inverse controller to track target vibration signals for the application to car subsystem evaluations. In recent times, perceptive assessments of car vibration have been technically significant, particularly parts interacting with passengers in the car such as steering wheels and seats. Conventional vibration test methods make it hard to track the target vibration signals in an accurate manner without compensating for the influence of the transfer function. Hence, this paper researched the vibration tracking system based on inverse system identification and digital signal processing technologies. Specifically, the controller employed a semi-active algorithm referring to both the offline modeling of the inverse system and the adaptive control. The semi-active controller could reconstruct the target vibration signal in a more efficient and safer way. The proposed methodology was first confirmed through computation simulations using Simulink. The simulation results verified that the semi-active controller could outperform the conventional active controller with respect to converging speed and stability. Following the simulation studies, actual vibration tests validated the suggested method in a steering wheel. A weight disturbance of about 0.24 kg was attached to the steering wheel to realize the possible change in the system characteristics. The semi-active controller could successfully track target vibration, such as a single or a dual harmonic signal, at the target spot of the steering wheel within the control error of about 1.6 dB regardless of the variation of the system transfer function. The proposed semi-active controller will provide an accurate, efficient tracking of vibrations in the evaluation of car subsystems.
The paper presents a robust adaptive control technique for precise regulation of a port fuel injection + direct injection (PFI+DI) system, a dual fuel injection configuration adopted in modern gasoline engines to boost performance, fuel efficiency, and emission reduction. Addressing parametric uncertainties on the actuators, inherent in complex fuel injection systems, the proposed approach utilizes an indirect model reference adaptive control scheme. To accommodate the increased control complexity in PFI+DI and the presence of additional uncertainties, a nonlinear plant model is employed, incorporating dynamics of the exhaust burned gas fraction. The primary objective is to optimize engine performance while minimizing fuel consumption and emissions in the presence of uncertainties. Stability and tracking performance of the adaptive controller are evaluated to ensure safe and reliable system operation under various conditions. Simulation studies demonstrate the reliability and effectiveness of the proposed control technique by subjecting it to aggressive time-varying uncertainties that emulate real-world scenarios with injector performance deviations. Results show that the proposed adaptive control maintains stable tracking performance even under aggressive injector uncertainties, showcasing its robustness in coping with varying environmental conditions and component behavior. This study contributes to the field of robust adaptive control for PFI+DI systems in gasoline engines, providing valuable insights into enhancing engine performance, fuel efficiency, and reducing emissions.
Aurora Flight Sciences, a Boeing Company Manassas, VA 703-369-3633
Vehicles-to-Everything or V2X communications provide attractive advantages in achieving reliable and high-performance connectivity amongst ground and aerial military vehicles. The 5G New Radio (NR) based cellular-V2X (C-V2X) technology, can support wide coverage areas with higher data rates and lower latencies needed for demanding military applications ranging from real-time sensing to navigation of autonomous military ground vehicles. Millimeter wave technology (mmWave) is critical to meet such throughput and latency requirements. However, mmWave links have a low transmission range and are often subject to blockages due to factors like weather, terrain, etc. that make them unreliable. Multi-connectivity with packet duplication can be used to enhance the reliability and latency by transmitting concurrently over independent links between a mobile device and multiple base stations. We propose and evaluate a novel method based on new radio dual connectivity (NR-DC) and packet duplication techniques to achieve reliable communication between military ground vehicles, especially in mobility scenarios. We further propose and analyze a Channel State Information Reference Signal Received Quality (CSI-RSRQ) based duplication strategy to improve the system's radio resource utilization. Channel State Information Reference Signal symbols in downlink transmissions are used to accurately compute the CSI-RSRQ values of the radio channel in real-time. This is critical on the battlefield for real-time awareness and adaptive control in fast-changing environments. Prototyped in the Simu5G network simulator and MATLAB, our results show packet duplication achieved less than 5 milliseconds of latency with zero packet loss under mobility.
An integrated electrically heated catalyst (EHC) in the three-way catalyst (TWC) of a gasoline internal combustion engine (ICE) is a promising technology to reduce engine cold-start pollutant emissions. Pre-heating the TWC ensures earlier catalyst light-off of a significant portion of the TWC. In such a case, the engine could readily be operated in a fuel-optimal manner since the engine cold-start emission is efficiently treated by the warmed-up EHC-equipped TWC. Pre-heating the EHC is an effective way to reduce cold-start emissions, among other possible EHC strategies. However, it might not always be possible to use pre-heating if the engine-start time is uncertain. In such a case, pre-heating can be started when the engine start is known with greater confidence and post-heating the catalyst could be followed. It would then be natural to turn off the EHC when the payoff for the electrical energy spent is no longer effective in engine cold-start emission reduction. The point in time at which to stop the EHC thus needs to be controlled. A model-free on-line adaptive controller aimed at minimizing the total equivalent emission is proposed, which is based on a set of pre-computed look-up optimum EHC stop times for the various possible fuel consumption trajectories. Compared to the theoretically optimal controller, the proposed controller gives a penalty of about 1% emission-based cost. A simulation framework for cold-start control and equivalent emission metric developed earlier are used in conjunction with a validation proposal to compare the performance of the candidate controllers.
Lateral control is an important part in the system of driverless mining trucks, which is used to realize accurate tracking of planned path. To solve the problem of poor accuracy of the existing single point preview algorithm, firstly, the lateral error model and the simplified truck dynamics model were built. The established truck dynamics model was verified and compared by simulation. The results show that the truck dynamic model in this paper retains accurate even at higher speed. Secondly, against the time delay of truck steering system, the cascade LQR-PID controller and MPC-MRAC controller are designed. The former resists the disturbance of steering time delay through the inner PID loop, while the latter realizes the adaptive control by establishing the steering model. Then, the dual-shift condition simulation was carried out by co-simulation model, and two controllers were compared and analyzed. The results show that the designed two controllers have good performance in the steering lag system. Finally, the LQR-PID controller was applied to the field test at speed 30km/h. The test results show that, compared with the single point preview algorithm, the LQR-PID controller obviously reduced the max tracking error from 1.125m to 0.792m and mean error from 0.644m to 0.236m, which achieved the desired effect.
Humanity has been interested in magnetism for over 300 years. Many authors have studied the use of applied magnetism to change the properties of products and expand the use of magnetic processing in ship repair production [1, 2]. Experience shows that magnetic pulse processing (MPP) is a simple and economical way to increase the durability of metal-cutting tools, increase the resource of the most worn parts of machines and mechanisms, and increase the durability of friction units, assembly units, and structures during their repair and manufacture. MPP has a number of advantages: simplicity of electromagnetic energy concentration on the product, its rapid accumulation by the material of the working elements of the part, and the efficiency of improving the operational characteristics (processing time is 0.3 ... 2.0 s with insignificant energy consumption). The indicated advantages of magnetic processing of products in comparison with other methods of hardening have been repeatedly confirmed by industrial experience in ship repair production [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]. However, the complexity of the practical application of MPP lies in the need to adjust the parameters of processing modes, taking into account a specific product. One of the most rational ways to solve this problem is to use adaptive control systems (CS). In the development and construction of such a system, it is advisable to use artificial intelligence methods in particular Adaptive Network-based Fuzzy Inference System networks (ANFIS), for solving optimization processing and forecasting MPP results. The use of these methods will allow for a modern approach to solving this problem. Therefore, the creation of an MPP installation with an intelligent control module is an urgent task today.
Items per page:
50
1 – 50 of 237