Search for: All records

The ubiquitous deployment of robots across diverse domains, from industrial automation to personal care, underscores their critical role in modern society. However, this growing dependence has also revealed security vulnerabilities. An attack vector involves the deployment of malicious software (malware) on robots, which can cause harm to robots themselves, users, and even the surrounding environment. Machine learning approaches, particularly supervised ones, have shown promise in malware detection by building intricate models to identify known malicious code patterns. However, these methods are inherently limited in detecting unseen or zero-day malware variants as they require regularly updated massive datasets that might be unavailable to robots. To address this challenge, we introduce ROBOGUARDZ, a novel malware detection framework based on zero-shot learning for robots. This approach allows ROBOGUARDZ to identify unseen malware by establishing relationships between known malicious code and benign behaviors, allowing detection even before the code executes on the robot. To ensure practical deployment in resource-constrained robotic hardware, we employ a unique parallel structured pruning and quantization strategy that compresses the ROBOGUARDZ detection model by 37.4% while maintaining its accuracy. This strategy reduces the size of the model and computational demands, making it suitable for real-world robotic systems. We evaluated ROBOGUARDZ on a recent dataset containing real-world binary executables from multi-sensor autonomous car controllers. The framework was deployed on two popular robot embedded hardware platforms. Our results demonstrate an average detection accuracy of 94.25% and a low false negative rate of 5.8% with a minimal latency of 20 ms, which demonstrates its effectiveness and practicality. 

Zero-day vulnerabilities pose a significant challenge to robot cyber-physical systems (CPS). Attackers can exploit software vulnerabilities in widely-used robotics software, such as the Robot Operating System (ROS), to manipulate robot behavior, compromising both safety and operational effectiveness. The hidden nature of these vulnerabilities requires strong defense mechanisms to guarantee the safety and dependability of robotic systems. In this paper, we introduce ROBOCOP, a cyber-physical attack detection framework designed to protect robots from zero-day threats. ROBOCOP leverages static software features in the pre-execution analysis along with runtime state monitoring to identify attack patterns and deviations that signal attacks, thus ensuring the robot’s operational integrity. We evaluated ROBOCOP on the F1-tenth autonomous car platform. It achieves a 93% detection accuracy against a variety of zero-day attacks targeting sensors, actuators, and controller logic. Importantly, in on-robot deployments, it identifies attacks in less than 7 seconds with a 12% computational overhead. 

Abstract Collaborative robots, or cobots, have been developed as a solution to the growing need for robots that can work alongside humans safely and effectively. One emerging technology in robotics is the use of Discrete Variable Stiffness Actuators (DVSAs), which enable robots to adjust their stiffness in a fast-discrete manner. This enables cobots to work in both low and high stiffness modes, allowing for safe collaboration with human workers or operation behind safety barriers. However, achieving good performance with different stiffness modes of DVSAs is a challenging problem. This paper proposes a method to provide force control of a DVSA by exploiting the dynamic model and the discrete stiffness levels. The two-mass dynamic model, a widely accepted model of flexible systems, is used to model and analyze the DVSA. The proposed method involves using Gain-scheduling and Deterministic Robust Control (DRC) controllers as modelbased control algorithms for the DVSA to achieve high-precision force control. We also conducted a comparison with the commonly used proportional integral derivative (PID) control algorithms. The paper presents a detailed analysis of the dynamic behavior of the DVSA and demonstrates the effectiveness of the proposed control algorithms through simulation with different scenario comparisons, even in the presence of external disturbances. 

Abstract This paper presents the development of a novel Actuation-Coordinated Mobile Parallel Robot (ACMPR), with a focus on studying the kinematics of the mobile parallel robot with three limbs (3-mPRS) comprising mobile prismatic joint-revolute joint-spherical joint. The objective of this research is to explore the feasibility and potential of utilizing omnidirectional mobile robots to construct a parallel mechanism with a mobile platform. To this end, a prototype of the 3-mPRS is built, and several experiments are conducted to identify the proposed kinematic parameters. The system identification of the 3-mPRS mobile parallel mechanism is conducted by analyzing the actuation inputs from the three mobile base robots. To track the motion of the robot, external devices such as the Vicon Camera are employed, and the data is fed through ROS. The collected data is processed based on the geometric properties, CAD design, and established kinematic equations in MATLAB, and the results are analyzed to evaluate the accuracy and effectiveness of the proposed calibration methods. The experiment results fall within the error range of the proposed calibration methods, indicating the successful identification of the system parameters. The differences between the measured values and the calculated values are further utilized to calibrate the 3-mPRS to better suit the experiment environment. 

Sensors in and around the environment becoming ubiquitous has ushered in the concept of smart animal agriculture which has the potential to greatly improve animal health and productivity using the concepts of remote health monitoring which is a necessity in times when there is a great demand for animal products. The data from in and around animals gathered from sensors dwelling in animal agriculture settings have made farms a part of the Internet of Things space. This has led to active research in developing efficient communication methodologies for farm networks. This study focuses on the first hop of any such farm network where the data from inside the body of the animals is to be communicated to a node dwelling outside the body of the animal. In this paper, we use novel experimental methods to calculate the channel loss of signal at sub-GHz frequencies of 100 - 900 MHz to characterize the in-body to out-of-body communication channel in large animals. A first-of-its-kind 3D bovine modeling is done with computer vision techniques for detailed morphological features of the animal body is used to perform Finite Element Method based Electromagnetic simulations. The results of the simulations are experimentally validated to come up with a complete channel modeling methodology for in-body to out-of-body animal body communication. The experimentally validated 3D bovine model is made available publicly on https://github.com/SparcLab/Bovine-FEM-Model.git GitHub. The results illustrate that an in-body to out-of-body communication channel is realizable from the rumen to the collar of ruminants with $$\leq {90}~{\rm dB}$$ path loss at sub-GHz frequencies ( $100-900~MHz$ ) making communication feasible. The developed methodology has been illustrated for ruminants but can also be used for other related in-body to out-of-body studies. Using the developed channel modeling technique, an efficient communication architecture can be formed for in-body to out-of-body communication in animals which paves the way for the design and development of future smart animal agriculture systems.