publications

2024

1. Topo-Geometrically Distinct Path Computation Using Neighborhood-Augmented Graph, and Its Application to Path Planning for a Tethered Robot in 3D

   Alp Sahin, and Subhrajit Bhattacharya

   IEEE Transactions on Robotics, 2024

   Abs Code

   Many robotics applications benefit from being able to compute multiple geodesic paths in a given configuration space. Existing paradigm is to use topological path planning, which can compute optimal paths in distinct topological classes. However, these methods usually require non-trivial geometric constructions which are prohibitively expensive in 3D, and are unable to distinguish between distinct topologically equivalent geodesics that are created due to high-cost/curvature regions or prismatic obstacles in 3D. In this paper, we propose an approach to compute k geodesic paths using the concept of a novel neighborhood-augmented graph, on which graph search algorithms can compute multiple optimal paths that are topo-geometrically distinct. Our approach does not require complex geometric constructions, and the resulting paths are not restricted to distinct topological classes, making the algorithm suitable for problems where finding and distinguishing between geodesic paths are of interest. We demonstrate the application of our algorithm to planning shortest traversible paths for a tethered robot in 3D with cable-length constraint.

2. Resonance Reduction Against Adversarial Attacks in Dynamic Networks via Eigenspectrum Optimization

   Alp Sahin, Nicolas Kozachuk, Rick S. Blum, and 1 more author

   2024

   Abs arXiv

   Resonance is a well-known phenomenon that happens in systems with second order dynamics. In this paper we address the fundamental question of making a network robust to signal being periodically pumped into it at or near a resonant frequency by an adversarial agent with the aim of saturating the network with the signal. Towards this goal, we develop the notion of network vulnerability, which is measured by the expected resonance amplitude on the network under a stochastically modeled adversarial attack. Assuming a second order dynamics model based on the network graph Laplacian matrix and a known stochastic model for the adversarial attack, we propose two methods for minimizing the network vulnerability that leverage the principle of eigenspectrum optimization. We provide extensive numerical results analyzing the effects of both methods.

3. Efficient Data Collection for Connected Vehicles With Embedded Feedback-Based Dynamic Feature Selection

   Zhuo Wang, Alp Sahin, and Xiangrui Zeng

   IEEE Transactions on Intelligent Vehicles, 2024

   Abs

   Collecting relevant and high-quality data is critical to machine-learning-based application development in automotive industry. It is highly desired to concentrate the connected vehicle data collection efforts on a set of useful data to avoid irrelevant or redundant data. For a given machine learning (ML) task, without a priori information, it can be challenging to determine what features to collect. In this work, we propose an efficient data collection workflow featuring a loop involving a connected vehicle fleet, a ML algorithm and a feedback-based dynamic feature selection strategy. The feature selection process is embedded in the data collection workflow, therefore it is more efficient than the traditional workflow where feature selection is applied after all data are collected. The proposed workflow reduces data transmission and storage cost. A feature selection algorithm that exploits and explores feature sets using an upper-confidence-bound-based method is designed for the trade-off between collecting known useful data and collecting potentially useful data. Validations are performed on two classification tasks to mimic the connected vehicle data collection: a synthetic problem based on a public ML dataset, and a robot ground surface classification problem with experiment data in a simulated data collection environment. The proposed method is tested against several baseline methods, including collecting all available data before feature selection, collecting random features, and running the data collection procedure episodically with embedded feature selection. Results show that the proposed algorithm in the closed-loop procedure achieves adequate performance with relatively few irrelevant and redundant data.

4. Real-Time Joystick-based Kinematics Control of a Snake-like Robot with Robustness Near Singular Configurations

   Alp Sahin, Stephen Brawner, Matt Bilsky, and 1 more author

   In 2024 IEEE Conference on Control Technology and Applications (CCTA), 2024

   Abs

   We propose algorithms for controlling an articulated snake-like robot using a 2-axis joystick. The key contributions of the paper are two-fold: i. Development of a pipeline for converting the joystick inputs into Cartesian velocity commands for the robot’s end-effector in a way that is intuitive to a user, and, ii. Development of an optimization-based controller that generates the necessary joint velocities to track the target Cartesian velocities as closely as possible even at or near singular configurations of the robot. We describe the proposed framework and outline its technical/mathematical details. We also demonstrate the effectiveness of the proposed approach through Gazebo simulations as well as real-robot experiments using a prototype of a snake-like robot called FLX BOT.

2023

1. Coordination-free Multi-robot Path Planning for Congestion Reduction Using Topological Reasoning

   Xiaolong Wang, Alp Sahin, and Subhrajit Bhattacharya

   Journal of Intelligent & Robotic Systems, Jul 2023

   Abs

   We consider the problem of multi-robot path planning in a complex, cluttered environment with the aim of reducing overall congestion in the environment, while avoiding any inter-robot communication or coordination. Such limitations may exist due to lack of communication or due to privacy restrictions (for example, autonomous vehicles may not want to share their locations or intents with other vehicles or even to a central server). The key insight that allows us to solve this problem is to stochastically distribute the robots across different routes in the environment by assigning them paths in different topologically distinct classes, so as to lower congestion and the overall travel time for all robots in the environment. We outline the computation of topologically distinct paths in a spatio-temporal configuration space and propose methods for the stochastic assignment of paths to the robots. A fast replanning algorithm and a potential field based controller allow robots to avoid collision with nearby agents while following the assigned path. Our simulation and experiment results show a significant advantage over shortest path following under such a coordination-free setup.

2021

1. Feedback-Based Dynamic Feature Selection for Constrained Continuous Data Acquisition

   Alp Sahin, and Xiangrui Zeng

   In 2021 American Control Conference (ACC), Jul 2021

   Abs

   Relevant and high-quality data are critical to successful development of machine learning applications. For machine learning applications on dynamic systems equipped with a large number of sensors, such as connected vehicles and robots, how to find relevant and high-quality data features in an efficient way is a challenging problem. In this work, we address the problem of feature selection in constrained continuous data acquisition. We propose a feedback-based dynamic feature selection algorithm that efficiently decides on the feature set for data collection from a dynamic system in a step-wise manner. We formulate the sequential feature selection procedure as a Markov Decision Process. The machine learning model performance feedback with an exploration component is used as the reward function in an ϵ-greedy action selection. Our evaluation shows that the proposed feedback-based feature selection algorithm has superior performance over constrained baseline methods and matching performance with unconstrained baseline methods.

2. Region-Based Planning for 3D Within-Hand-Manipulation via Variable Friction Robot Fingers and Extrinsic Contacts

   Alp Sahin, Adam J. Spiers, and Berk Calli

   In 2021 IEEE International Conference on Robotics and Automation (ICRA), Jul 2021

   Abs

   Attempts to achieve robotic Within-Hand-Manipulation (WIHM) generally utilize either high-DOF robotic hands with elaborate sensing apparatus or multi-arm robotic systems. In prior work we presented a simple robot hand with variable friction robot fingers, which allow a low-complexity approach to within-hand object translation and rotation, though this manipulation was limited to planar actions. In this work we extend the capabilities of this system to 3D manipulation with a novel region-based WIHM planning algorithm and utilizing extrinsic contacts. The ability to modulate finger friction enhances extrinsic dexterity for three-dimensional WIHM, and allows us to operate in the quasi-static level. The region-based planner automatically generates 3D manipulation sequences with a modified A* formulation that navigates the contact regions between the fingers and the object surface to reach desired regions. Central to this method is a set of object-motion primitives (i.e. within-hand sliding, rotation and pivoting), which can easily be achieved via changing contact friction. A wide range of goal regions can be achieved via this approach, which is demonstrated via real robot experiments following a standardized in-hand manipulation benchmarking protocol.

3. A Framework for Benchmarking Feedback-Based Dynamic Data Collection Methods in Connected Vehicle Networks

   Alp Sahin, and Xiangrui Zeng

   In SAE WCX Digital Summit, Apr 2021

   Abs

   Supervised, unsupervised and active learning techniques can be used to develop prognostics and diagnostics in connected vehicle networks, where a wide variety of sensors are available for data collection. However, constraints placed by the on-board equipment, vehicle network, time, and human resources limit the amount of sensor data and labels for machine learning. When no prior information about the data distribution or domain knowledge is available, it becomes a challenging task to collect limited and relevant data to train a machine learning model matching the desired performance threshold. To tackle this challenge, techniques such as experimental design, feature selection, and active learning can be applied, and the data collection process can be advanced to a closed-loop system where new data collection decisions are made based on the feedback from collected data. In this paper, an iterative design and evaluation procedure is considered to develop and deploy these feedback-based decision-making algorithms. So far, no simulation-based benchmarking frameworks are available for evaluating and testing the performance of these methods in a connected vehicle network setting that can impose the real-world conditions. In this work, we propose a simulation platform that makes use of static datasets to simulate dynamic data collection process and provide comprehensive evaluations of feedback-based dynamic data collection decision-making algorithms. The platform provide means for stepwise input of desired dynamic data collection decisions and delayed return of corresponding data to mimic the real-world data collection procedure. Compliant experimental design, feature selection, and active learning strategies can be used within the framework to determine the data to be collected at each step. We also provide an implementation example of a wrapper-based feature selection algorithm using greedy search and exploration components along with a random feature selector for comparative analysis.