home it section

Deep Learning Flight Control

Real-time implementation of nonlinear model predictive control for high angle of attack Maneuvers in fighter aircrafts using deep learning

Nonlinear model predictive control (NMPC) provides a nice framework for accounting for multivariable nonlinear dynamics subject to constraints; however, the cost of implementing NMPC is high due to the need to solve a large-scale nonlinear program to optimality in real-time. This research is focussed on the design of real-time solution to NMPC (low computations) for fast dynamic systems. In this work, a comprehensive and innovative solution based on a deep learning-based approximate NMPC scheme has been proposed to overcome these computational challenges. The main idea is to learn a deep neural network followed by the symbolic simplification of the NMPC law, whose evaluation cost and memory footprint can be optimized offline. Since not all the states of the system are measured, we also combine the deep learning based NMPC approach with an unscented Kalman filter (UKF) as well as show how this scheme can be easily modified to be offset-free (i.e. remove steady-state error due to persistent disturbances and modelling error). The proposed approach is implemented on a highly fast dynamic system (i.e. F-16 fighter aircraft to perform a high angle of attack maneuver) to check the controller efficiency. Fighter aircraft need to be able to make fast and complex maneuvers (that exploit strongly nonlinear dynamics) to be able to maintain air superiority. However, such aggressive maneuvers require strong coordination between the actions to avoid destabilizing the system. A quantitative analysis from comparison with the standard MPC approach shows the practical utility of the proposed approach.

https://www.tandfonline.com/doi/full/10.1080/21642583.2024.2342819

high-voltage

Autonomous UAV Landing Control

Autonomous Landing of an UAV Using H∞ Based Model Predictive Control

Possibly the most critical phase of an Unmanned Air Vehicle (UAV) flight is landing. To reduce the risk due to pilot error, autonomous landing systems can be used. Environmental disturbances such as wind shear can jeopardize safe landing, therefore a well-adjusted and robust control system is required to maintain the performance requirements during landing. The paper proposes a loop-shaping-based Model Predictive Control (MPC) approach for autonomous UAV landings. Instead of conventional MPC plant model augmentation, the input and output weights are designed in the frequency domain to meet the transient and steady-state performance requirements. Then, the ℋ∞ loop shaping design procedure is used to synthesize the state-feedback controller for the shaped plant. This linear state-feedback control law is then used to solve an inverse optimization problem to design the cost function matrices for MPC. The designed MPC inherits the small-signal characteristics of the 𝐻∞ controller when constraints are inactive (i.e., perturbation around equilibrium points that keep the system within saturation limits). The ℋ∞ loop shaping synthesis results in an observer plus state feedback structure. This state estimator initializes the MPC problem at each time step. The control law is successfully evaluated in a non-linear simulation environment under moderate and severe wind downburst. It rejects unmeasured disturbances, has good transient performance, provides an excellent stability margin, and enforces input constraints.

https://www.mdpi.com/2504-446X/6/12/416

sec

Connectivity-Driven Energy Optimization

Energy economization using connectivity-based eco-routing and driving for fleet of battery electric vehicles

Energy economizing using Connectivity-based eco-Routing and Driving (CeRD) is a key to conserve the resources of a fleet. This work takes maximum advantage by advising optimal energy-consumption-based route and velocity. Product delivery to several universities by a group of battery electric vehicles (BEVs) is considered. Problem contains certain constraints, i.e., BEV load capacity, time window, resources restriction, speed bounds, battery charging/discharging limits, and traffic. Simulation of urban mobility (SUMO) is used for the estimation of traffic and routes. Selection of route is based on energy consumption of traffic’s velocity. Dijkstra’s algorithm is applied to select the route with least energy consumption, and Clarke and Wright (C&W) algorithm is used to find the optimal sequence of visiting universities. Pontryagin’s minimum principle (PMP) is then applied to find the optimal velocity profiles for BEVs on the optimized routes. Results have shown 60%–70% reduction in energy consumption along with improvement in charging rates when compared with the vehicles without CeRD assistance in different scenarios. Furthermore, the results are validated for distance and time using online route planner, for energy efficiency using electric vehicles’ database, and for energy consumptions using a driver model. CeRD has also improved 30% route distance and 40% trip time.

https://ieeexplore.ieee.org/abstract/document/10143086

n-ev-retrofits

Intelligent Multi-Vehicle Planning

Optimizing Energy and Battery Health Using Route, Allocation and Velocity Planning for a Multi Electric Vehicle System

The rising need for green transport drives the imperative to refine electric vehicle (EV) operations, prioritizing both single charge range and battery lifespan. A comprehensive approach is presented to enhance the range of EVs and battery life by optimizing energy consumption and charging/discharging behavior. The problem focuses on the selection of energy-efficient routes, customer allocation, and driving speed for a multi-EV system. Constraints include: depth and rate of charging/discharging, EVs’ capacity and quantity, customers’ availability and demand, speed and acceleration limits, bounds on input torque and its rate of change, and traffic signals, while uncertainties cover slippery road surface and grade. Dijkstra’s algorithm, mixed integer linear programming (MILP) algorithm, and model predictive control (MPC) algorithm are used to perform the task. Energy consumption improved by 6.6%, when compared with integrated routing and driving technique using Clarke and Wright (C&W) and Pontryagin’s minimum principle (PMP), respectively. Battery effective capacity and lifespan increase from 27 to 31.2 kWh and 7.12 to 8.85 years, respectively. The improvement in range per charge and over lifespan compared with commercially available data of Dacia Spring Electric 45 and Leapmotor T03 further strengthens the claim. Almost the same execution time is obtained for the proposed integrated strategy when compared with the standalone routing and driving techniques.

https://ieeexplore.ieee.org/abstract/document/11096722

n-ept

Machine-Driven Interaction Discovery

Large scale biological interaction network discovery through knowledge graph driven machine learning

Biological systems derive from complex interactions between entities ranging from biomolecules to macroscopic structures, forming intricate networks essential for understanding disease mechanisms and developing therapeutic interventions. Current AI-driven interaction predictors typically operate in isolation, focusing on single tasks and missing the broader picture of how different biological interactions influence each other. Traditional wet-lab approaches for identifying these interactions are expensive, time-consuming, and error-prone. No unified platform currently exists where biologists can predict and analyze multiple types of biological relationships comprehensively, limiting our ability to discover new therapeutic applications and fully understand interconnected biological mechanisms.

https://link.springer.com/article/10.1186/s12967-025-06789-5

pharma

Sliding-Mode Glider Guidance

Range guidance for subsonic unpowered gliding vehicle using integral action-based sliding mode control

Range enhancement for a Subsonic Unpowered Gliding Vehicle (SUGV) is an interesting and challenging problem because it depends only on mechanically stored energy. In this study, the inverted Y-tail joint stand-off weapon is selected as the SUGV. To increase the range, a range guidance scheme has been developed which depends on the dispersion points and the angle of attack. To overcome saturation effect in range guidance command and nonsmooth range guidance command that cause a terminal error in range, an Integral Action-based Sliding Mode Control (IA SMC) methodology is proposed. The stability analysis of the IA SMC is also established by the quadratic Lyapunov function. To deal with the chattering problem in SMC and enhance the velocity of reaching phase, a tan−1-based strong exponential reaching law is also employed. When compared with the PID-based sliding surface dependent SMC (PID SMC) to achieve a range of 125 km, MATLAB simulation results show that IA SMC generates smooth guidance command and reduces the saturation effects and terminal error in the range. Furthermore, the quantitative analysis shows that the effectiveness of IA SMC is better than that of PID SMC.

Learn More

Automation

Longitudinal Sliding Mode Control

Longitudinal Cruise Control of a Car using Sliding Mode Approach

From the last few years, there has occur an increase in number of cars, which has badly affected our transport system and specially traffic on the roads. Traffic congestion and traffic accidents are the major concerns due to an increase in number of vehicles. Many efforts have been made in the last few years to improve safety and comfort in vehicular system. One of present focus is intelligent transport system (ITS) to ensure safety and comfort for the passenger. The main focus of this paper is to control the longitudinal dynamics of a vehicle, which make sure that a safe distance is kept between the leading and following vehicle. The major drawback with the existing control techniques is the inability to perform well in the presence of various parameters, which vary with the variation in the external factors like temperature and etc. This paper focuses on the development of a sliding mode control (SMC) for the longitudinal dynamics of automotive vehicle in a platoon system. System response will be observed in the presence of parametric uncertainty. First order and Second order sliding mode is applied and results are compared by comparing chattering in both the cases.

Learn More
Torq

Optimized Genetic Powertrain Controller

Genetic algorithms optimized multi-objective controller for an induction machine based electrified powertrain

In electrified powertrain control, meeting the torque demands and ensuring efficient Electrical Machine (EM) operations are two essential but conflicting demands. A multi-objective Linear Parameters Varying (LPV) controller is proposed to address the problem of these conflicting objectives. The synthesis of multi-objective controller is based on the selection of optimal weighting functions optimized by Genetic Algorithm (GA). The effectiveness of the proposed controller is tested and evaluated for an electrified powertrain operating in a standard urban driving cycles. The stability of the proposed Multi-Objective Controller (MOC) is established. The nonlinear simulation of the proposed controller delivers the robust performance and better efficiency of an EV Induction Machine (IM) based electric drive over the entire driving cycle.

Learn More
EV

Guidance of air vehicles

A sliding mode approach

This paper presents a novel nonlinear guidance scheme for ground track control of aerial vehicles. The proposed guidance logic is derived using the sliding mode control technique, and is particularly suited for unmanned aerial vehicle (UAV) applications. The main objective of the guidance algorithm is to control the lateral track error of the vehicle during flight, and to keep it as small as possible. This is achieved by banking the vehicle, that is, by executing roll maneuvers. The guidance scheme must perform well both for small and large lateral track errors, without saturating the roll angle of the vehicle, which serves as the control input for the guidance algorithm. The limitations of a linear sliding surface for lateral guidance are indicated; a nonlinear sliding surface is thereafter proposed which overcomes these limitations, and also meets the criterion of a good helmsman. Stability of the nonlinear surface is proved using Lyapunov theory; control boundedness is also proved to ensure that the controls are not saturated even for large track errors. The proposed guidance law is implemented on the flight control computer of a scaled YAK-54 UAV and flight results for different scenarios (consisting of both small and large errors) are presented and discussed. The flight test results confirm the effectiveness and robustness of the proposed guidance scheme.

Learn More