More Like this

1. Counter-Misdirection in Behavior-Based Multi-Robot Teams

   Chen, Shengkang; Arkin, Ronald C. (March 2021, 2021 IEEE International Conference on Intelligence and Safety for Robotics)

   null (Ed.)

   When teams of mobile robots are tasked with different goals in a competitive environment, misdirection and counter-misdirection can provide significant advantages. Researchers have studied different misdirection methods but the number of approaches on counter-misdirection for multi-robot systems is still limited. In this work, a novel counter-misdirection approach for behavior-based multi-robot teams is developed by deploying a new type of agent: counter misdirection agents (CMAs). These agents can detect the misdirection process and “push back” the misdirected agents collaboratively to stop the misdirection process. This approach has been implemented not only in simulation for various conditions, but also on a physical robotic testbed to study its effectiveness. It shows that this approach can stop the misdirection process effectively with a sufficient number of CMAs. This novel counter-misdirection approach can potentially be applied to different competitive scenarios such as military and sports applications. 

   more » « less
2. Push and Pull: Shepherding Multi-Agent Robot Teams in Adversarial Situations

   https://doi.org/10.1109/ARSO46408.2019.8948797  

   Pettinati, Michael J.; Arkin, Ronald C. (October 2019, Proceedings of 15th IEEE International Conference on Advanced Robotics and Its SOcial Impacts (ARSO-2019))

   Teams of robots tasked with making critical decisions in competitive environments are at risk for being shepherded or misdirected to a location that is advantageous for a competing team. Our lab is working to understand how adversarial teams of robots can successfully move their competition to desired locations in part so that we can then devise practices to counter these strategies and help make team functioning more successful and secure. In this paper, preliminary research is presented that studies how a team of robots can be shepherded or misdirected to a disadvantageous location. We draw inspiration from herding practices as well as deceptive practices seen in higher-order primates and humans. We define behaviors for the target (mark) agents to be moved as well as members of the shepherding team (a pushing agent and pulling shills) and present simulation results showing how these behaviors move robots to a desired location. These behaviors were implemented and trialed on hardware platform. A discussion of ongoing research into understanding misdirection in multi-robot teams concludes this paper. 

   more » « less
3. Misdirection in Robot Teams: Methods and Ethical Considerations

   Arkin, Ronald C. (April 2019, 2019 International Association for Computing and Philosophy annual conference)

   Trust, dependability, cohesion, and capability are integral to an effective team. These attributes are the same for teams of robots. When multiple teams with competing incentives are tasked, a strategy, if available, may be to weaken, influence or sway the attributes of other teams and limit their understanding of their full range of options. Such strategies are widely found in nature and in sporting contests such as feints, misdirection, etc. This talk focuses on one class of higher-level strategies for multi-robots, i.e., to intentionally misdirect using shills or confederates where needed, and the ethical considerations associated with deploying such teams. As multi-robot systems become more autonomous, distributed, networked, numerous, and with more capability to make critical decisions, the prospect for intentional and unintentional misdirection must be anticipated. While benefits are clearly apparent to the team performing the deception, ethical questions surrounding the use of misdirection or other forms of deception are quite real. 

   more » « less
4. RoboGuardZ: A Scalable Zero-Shot Framework for Detecting Zero-Day Malware in Robots

   Kaur, Upinder; Celik, Z Berkay; Voyles, Richard M (October 2024, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS))

   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. 

   more » « less
5. Level Curve Tracking without Localization Enabled by Recurrent Neural Networks

   https://doi.org/10.1109/CACRE50138.2020.9230272  

   Zhang, Ziqiao; Al-Abri, Said; Wu, Wencen; Zhang, Fumin (September 2020, 2020 5th International Conference on Automation, Control and Robotics Engineering (CACRE))

   null (Ed.)

   Recursive neural networks can be trained to serve as a memory for robots to perform intelligent behaviors when localization is not available. This paper develops an approach to convert a spatial map, represented as a scalar field, into a trained memory represented by the long short-term memory (LSTM) neural network. The trained memory can be retrieved through sensor measurements collected by robots to achieve intelligent behaviors, such as tracking level curves in the map. Memory retrieval does not require robot locations. The retrieved information is combined with sensor measurements through a Kalman filter enabled by the LSTM (LSTM-KF). Furthermore, a level curve tracking control law is designed. Simulation results show that the LSTM-KF and the control law are effective to generate level curve tracking behaviors for single-robot and multi-robot teams. 

   more » « less