Manish Kumar

University of Cincinnati | UC · School of Dynamic Systems

Central Pattern Generators (CPGs) form the neural basis of the observed rhythmic behaviors for locomotion in legged animals. The CPG dynamics organized into networks allow the emergence of complex locomotor behaviors. In this work, we take this inspiration for developing walking behaviors in multi-legged robots. We present novel DeepCPG policies th...

Central Pattern Generators (CPGs) form the neural basis of the observed rhythmic behaviors for locomotion in legged animals. The CPG dynamics organized into networks allow the emergence of complex locomotor behaviors. In this work, we take this inspiration for developing walking behaviors in multi-legged robots. We present novel DeepCPG policies th...

Hybrid UAV has combined characteristics of fixed-wing UAV and multi-rotor UAVs. Fixed-wing UAV has higher flight endurance, while the multi-rotor UAV requires no runways and can also perform vertical take-off and landing. Hybrid morphing wing UAV has ability to perform mission both as quadcopter and fixed-wing aircraft combined in efficient manner....

In this paper, the design, development, and flight testing of the technology demonstration prototype for the sliding arm quadcopter are described. Furthermore, details on the hardware component design, machining, and integration of various UAV subsystems are discussed. The sliding arm UAV is a shape changing aerial robot design that has additional...

Mixed Integer Linear Programs (MILPs) are often used in the path planning of both ground and aerial vehicles. Such a formulation of the path planning problem requires a linear objective function and constraints, limiting the fidelity of the the tracking of vehicle states. One such parameter is the state of charge of the battery used to power the ve...

This paper proposes a sliding mode controller for a novel kind of quadcopter called tilt-rotor quadcopter. The controller is modeled in SIMULINK and implemented on a Pixhawk 2 flight controller board. First, this paper presents the dynamic model of the tilt-rotor quadcopter. Second, the proposed smart sliding mode controller for simulation studies...

Deep reinforcement learning (RL) has made it possible to solve complex robotics problems using neural networks as function approximators. However, the policies trained on stationary environments suffer in terms of generalization when transferred from one environment to another. In this work, we use Robust Markov Decision Processes (RMDP) to train t...

In this paper, we present the modeling and control of a tethered tilt-rotor quadcopter (TRQ). They can achieve long-endurance flights as the system is powered continuously using a taut power cable. The conventional quadcopters have already been utilized with tether applications. However, the tilt-rotor quadcopter has additional servo motors for ach...

In this paper, an interacting multiple model (IMM) is proposed to generate future position estimates and enable the safe deployment of small unmanned aerial systems (sUAS) in the national airspace (NAS). The proposed system has been developed in the ROS-Gazebo physics engine environment. The IMM module is executed on the ground control station (GCS...

We investigate the spatiotemporal dynamics and control of an epidemic using a partial differential equation (PDE) based Susceptible-Latent-Infected-Recovered (SLIR) model. We first validate the model using empirical COVID−19 data corresponding to a period of 45 days from the state of Ohio, United States. Upon optimizing the model parameters in the...

Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) have been used in research and development community due to their strong potential in high-risk missions. One of the most important civilian implementations of UAV/UGV cooperative path planning is delivering medical or emergency supplies during disasters such as wildfires, the focu...

In this paper, we present the implementation of a multiple model system to generate future position estimates and enable safe deployment of small unmanned aerial systems (sUAS) in the national airspace (NAS). The system was developed in the ROS-Gazebo physics engine environment. The proposed system was developed for multi-rotor and fixed wing simul...

Different approaches to Visual-Inertial odometry(VIO) has been presented in literature. However, few research works exploits the quadrotor translational and rotational dynamics and the known thrust and torque inputs. Additional information from the dynamics with known control inputs improves the state estimation. This paper is focused on the estima...

In this paper, we present the modeling and control of a tethered tilt-rotor quadcopter (TRQ). They can achieve long-endurance flights as the system is powered continuously using a taut power cable. The conventional quadcopters have already been utilized with tether applications. However, the tilt-rotor quadcopter has additional servo motors for ach...

A fuzzy logic (FL) force-torque feedback controller (FT-FC) is utilised to solve the problem of multi-drone cooperative transport with known offset center of gravity (CG). The control scheme is decentralised in the sense that no inter drone communication is required. The controller only needs measurements of contact forces and torques and minimum o...

Ants demonstrate effective team effort and heavy lift capability during collaborative transportation (CT) of their food. In such a biological system, each ant acts independently and knows how to comply with it’s team. The work presented in this paper attempts to mimic simple ant strategies, with minimal assumptions (thus termed as an Ockham’s razor...

We study the spatiotemporal dynamics of an epidemic spread using a compartmentalized PDE model. The model is validated using COVID-19 data from Hamilton County, Ohio, USA. The model parameters are estimated using a month of recorded data and then used to forecast the infection spread over the next ten days. The model is able to accurately estimate...

Deep reinforcement learning (RL) has made it possible to solve complex robotics problems using neural networks as function approximators. However, the policies trained on stationary environments suffer in terms of generalization when transferred from one environment to another. In this work, we use Robust Markov Decision Processes (RMDP) to train t...

Link to codes and data used in this work.

In this paper, we present a novel developmental reinforcement learning-based controller for a quadcopter with thrust vectoring capabilities. This multirotor UAV design has tilt-enabled rotors. It utilizes the rotor force magnitude and direction to achieve the desired state during flight. The control policy of this robot is learned using the policy...

Tethered drone systems can be used to perform long-endurance tasks such as area surveillance and relay stations for wireless communication. However, all the existing systems use tethers only for data and power transmission from a stationary point on the ground. This work presents a control strategy that enables a quadcopter to follow a moving tethe...

In this study we consider a Dynamic Genetic Algorithm used to optimize the movement of a symmetric six-legged creature. The optimal movement is that which advances the creature in a straight line forward with the greatest average speed. The mutation rate and crossover rate are adjusted based on number of iterations the algorithm has completed. This...

We investigate ant-colony-inspired foraging strategies for enhancing the efficiency of a swarm of artificial agents engaged in a search-and-retrieval application. First, we extend a mathematical model of ant foraging to account for the evolution of the information collected during search-and-retrieval over time. We then use the extended model to nu...

In this paper, we present a controller design for a quadrotor by obtaining the derivative of the actual control input using the concept of multiple sliding surfaces and Lyapunov stability analysis. The conventional sliding mode controller is highly robust. The discontinuous part of the control input suppresses disturbances well. Theoretically, howe...

Proportional-Derivative (PD) controllers are commonly used in quadrotors due to their simple structure. Tuning of the gains of the PD controller is often cumbersome due to strong coupling of the dynamics between three linear and three angular degrees of freedom. This paper presents a novel method of auto adjusting the proportional and derivative ga...

In this work, a fuzzy inference system (FIS) is proposed to develop a sense and avoid technique for conflict-free UAV flight operations in the national airspace. The proposed method is implemented alongside the flight control software guiding the UAV towards a predefined goal. The fuzzy system makes decisions for sense and avoid with respect to the...

In this work, a Fuzzy Inference System (FIS) is proposed to develop a sense and avoid technique for conflict-free Unmanned Aerial Vehicle (UAV) flight operations in the national airspace. The proposed method is implemented alongside the flight control software guiding the UAV towards a predefined goal. The fuzzy system makes decisions for sense and...

In this paper, we present a novel developmental reinforcement learning-based controller for a quadcopter with thrust vectoring capabilities. This multirotor UAV design has tilt-enabled rotors. It utilizes the rotor force magnitude and direction to achieve the desired state during flight. The control policy of this robot is learned using the policy...

In this paper, we present an autonomous flight controller for a quadcopter with thrust vectoring capabilities. This UAV falls in the category of multirotors with tilt-motion enabled rotors. Since the vehicle considered is over-actuated in nature, the dynamics and control allocation have to be analysed carefully. Moreover, the possibility of hoverin...

Recently, there has been a tremendous increase of interest in utilizing Unmanned Aerial Vehicles (UAVs) for a number of civilian applications. With this increased interest, it is imperative that these UAVs are able to operate in shared airspace for enhanced efficiency. Multi-UAV systems are inherently safety-critical systems, which means that safet...

The conceptual design and flight controller of a novel kind of quadcopter are presented. This design is capable of morphing the shape of the UAV during flight to achieve position and attitude control. We consider a dynamic center of gravity (CoG) which causes continuous variation in a moment of inertia (MoI) parameters of the UAV in this design. Th...

Multi-rotor Platform Based UAV Systems provides an excellent opportunity for experiential learning, capability augmentation and confidence-building for senior level undergraduates, entry-level graduates, engineers working in government agencies, and industry involved in UAV R&D. Topics in this book include an introduction to VTOL multi-copter UAV p...

Digitization has led to smart, connected technologies be an integral part of businesses, governments and communities. For manufacturing digitization, there has been active research and development with a focus on Cloud Manufacturing (CM) and the Industrial Internet of Things (IIoT). This work presents a computer vision toolkit (CV Toolkit) for non-...

Quality control is an essential process in manufacturing to make the product defect-free as well as to meet customer needs. The automation of this process is important to maintain high quality along with the high manufacturing throughput. With recent developments in deep learning and computer vision technologies, it has become possible to detect va...

This paper proposes a novel bio-inspired method for Unmanned Aerial Vehicle (UAV) trajectory planning in three dimensions using Partial Differential Equations (PDEs) for application in dynamic hostile environments. The proposed method exploits the dynamical property of fluid flowing through a porous medium. This method evaluates risk to generate po...

Cooperative transportation by multiple Unmanned Aerial Vehicles (UAVs) has been a topic of interest amongst robotics researchers since a decade. Researchers have developed different control schemes to address some of the issues related to cooperative transport. However, most of the existing control strategies assume a stationary center of gravity (...

Unmanned Aerial Vehicles (UAVs) have great potential in civilian applications, such as package delivery, agriculture, and disaster management. The number of UAVs in these applications will continue to increase, and thus so does the importance of safely integrating them into the National Airspace System (NAS), with one key component being the manage...

Small unmanned aerial vehicles (UAVs) have the potential to revolutionize various applications in civilian domain such as disaster management, search and rescue operations, law enforcement, precision agriculture, and package delivery. As the number of such UAVs rise, a robust and reliable traffic management is needed for their integration in nation...

Human motion is highly variable due to the effects of interaction with the environment and the intentionality of movement. Current assistive device control systems struggle to support this variable human motion as they restrict typically uncontrolled degrees of freedom resulting in unnatural, mechanistic and highly repetitive (i.e., artificially in...

In this paper, fault-tolerance characteristics of a reconfigurable tilt-rotor quadcopter upon a propeller failure are presented. Traditional quadcopters experience instability and asymmetry about yaw-axis upon a propeller failure but the design and control strategy presented here can handle a complete propeller failure during flight. Fault-toleranc...

Tilt-rotor quadcopters are a novel class of quadcopters with a servo motor attached on each arm that assist the quadcopter's rotors to tilt to a desired angle thereby enabling thrust vector-ing. Using these additional tilt angles, this type of a quadcopter can be used to achieve desired trajectories with faster maneuvering and can handle external d...

In this paper, we address the decentralized collaborative trajectory planning and target surrounding of multiple Unmanned Air Vehicles (UAVs) in three-dimensional space using Partial Differential Equation (PDE) method. The mission objective is simultaneously arrival of UAVs with safe flight trajectory to a certain radius of an a – priori target. Th...

Multi-robot systems have an innate advantage of enhancing system robustness in situations where the chances for noise in sensory data or faults in robotic agents are high. The distribution of the task at hand among multiple robots and coordination among these robotic agents can also make these systems more efficient. The formation problem for a mul...

FlyMASTER is a software platform developed to advance research in the area of collaborative multi-agent UAV systems. FlyMASTER provides researchers of multi-agent UAV systems with a software platform promoting ease of integration, rapid development and flexibility to further advancement in the field of cooperative swarming UAVs. Its design, coupled...

The ability to move one’s body from sitting to standing is a crucial ability for independent living. Especially for seniors with decreasing muscular strength, sit-to-stand (STS) transitions are exceptionally risky and often call for assistance. In general, an STS transition is a complex full-body activity that requires the synergistic coordination...

Unmanned Air Vehicles (UAVs), which have been popular in the military context, have recently attracted attention of many researchers because of their potential civilian applications. However, before UAVs can fly in civilian airspace, they need to be able to navigate safely to their goal while maintaining separation with other manned and unmanned ai...

Unmanned aerial vehicles (UAVs) have recently attracted the attention of researchers due to their numerous potential civilian applications. However, current robot navigation technologies need further development for efficient application to various scenarios. One key issue is the "Sense and Avoid" capability, currently of immense interest to resear...

Networks of unmanned aerial vehicles (UAVs), capable of providing flexible aerial views over large areas, are playing important roles in today's distributed sensing systems. Since camera sensors are sensitive to occlusions, it is more challenging for UAVs to provide satisfactory sensing quality in geometrically complex environments, such as dense u...

In this paper, the control maneuvering, and performance analysis of a tilting-rotor quadcopter during autonomous flight is presented. Unlike traditional quadcopters, a tilting-rotor quad-copter provides additional actuated controls as the propeller motors are actuated for tilt which can be utilized to improve efficiency of the aerial vehicle during...

In this paper, we present a feed-forward control approach for complex trajectory tracking by a tilting-rotor quadcopter during autonomous flight. Tilting-rotor quadcopter is a more agile version of conventional quadcopter as the propeller motors are actuated to tilt about the quadcopter arm. The tilt-rotor quad-copter is capable of following comple...

A non-linear control of a tilt-rotor quadcopter using sliding mode technique is presented in this paper. The tilt-rotor quad-copters are a novel class of quadcopters with a servo motor installed on each arm that enables the quadcopter's rotors to tilt to a particular angle. Using these additional tilt angles, this type of a quadcopter can be used t...

Approximately 1.5 million senior citizens live under nursing supervision, and most require assistance with at least one or more Activities of Daily Living (ADL). These include transferring in and out of chairs, beds and toilets, which necessitates the ability to perform sit-to-stand transitions. The sit-to-stand transition is a complex full-body ac...

This paper presents a novel control approach to perform collaborative transportation by using multiple quadcopter Unmanned Aerial Vehicles (UAVs). In this paper, a leader-follower approach is implemented. The leader UAV uses a Proportional, Integral and Derivative (PID) controller to reach the desired goal point or follow a predefined trajectory. T...

Design of robot swarms inspired by self-organization in social insect groups is currently an active research area with a diverse portfolio of potential applications. In this work, the authors propose a control law for efficient area coverage by a robot swarm in a 2D spatial domain, inspired by the unique dynamical characteristics of ant foraging. T...

Population of the world above the age of 65 years is increasing rapidly. Aging causes weakening of human joints which increases constraints on mobility of the body. Sit-to-Stand (STS), an important part of Activities of Daily Living (ADL) is one of the motions that is affected because of weakened joints. With the lack of personal care there is goin...

This paper proposes a fast method for obtaining mathematically optimal trajectories for UAVs while avoiding collisions. A comparison of the proposed method with previously used Mixed Integer Linear Programming (MILP) to find the optimal collision-free path UAVs, aircraft, and spacecraft show the effectiveness and performance of this method. Here, t...

In this paper, we present a differential flatness based hybrid flight controller for the quadcopter UAV. The combination of conventional PID based controller with the full state feedback based LQR controller results in the proposed hybrid controller. The performance of the resulting controller is further enhanced by using differential flatness base...

Biomimetic robots have gained attention recently for various applications ranging from resource hunting to search and rescue operations during disasters. Biological species are known to intuitively learn from the environment, gather and process data, and make appropriate decisions. Such sophisticated computing capabilities in robots are difficult t...

We recently developed a novel learning solution for unsupervised learning in robots based on resistive memory devices arranged in a crossbar fashion and validated it by navigating a robot in an unknown environment with randomly placed obstacles. In this work, we study the effects of variations in device doping concentrations and the resistive state...

This work proposes a control law for efficient area coverage and pop-up threat detection by a robot swarm inspired by the dynamical behavior of ant colonies foraging for food. In the first part, performance metrics that evaluate area coverage in terms of characteristics such as rate, completeness and frequency of coverage are developed. Next, the K...

The purpose of this work is to demonstrate the efficacy of using a fuzzy logic based approach to successfully perform the landing maneuver of a small multirotor aircraft on a dynamic target at low speeds. The corrective control of the vehicle is exerted based on visual input data from an on-board camera system. Given the constraint of the vehicle s...