Precision agriculture applications, such as real-time crop monitoring via robots for pest and disease detection, need ultra-low latency and high-definition video streaming to enable timely decision-making and intervention. Although 5G networks have promised low latency, current streaming latency over challenging rural networks can hinder timely agricultural action. This paper suggests an adaptive solution to reduce latency in delay-sensitive precision agriculture by integrating deep reinforcement learning (DRL) with semantic communication techniques. We propose a system that replaces traditional compression methods with a modified MambaJSCC framework, focusing on the region of interest (ROI) in agricultural imagery. Our experimental results show that our method achieves significantly lower streaming latency, potentially by up to 56%, along with high-quality frame reconstruction in these agriculturally relevant regions. Our work presents a technically viable solution for enhancing real-time, delay-sensitive agricultural monitoring applications.

A sudden roadblock on highways due to many reasons such as road maintenance, accidents, and car repair is a common situation we encounter almost daily. Autonomous Vehicles (AVs) equipped with sensors that can acquire vehicle dynamics such as speed, acceleration, and location can make intelligent decisions to change lanes before reaching a roadblock. A number of literature studies have examined car-following models and lane-changing models. However, only a few studies proposed an integrated car-following and lane-changing model, which has the potential to model practical driving maneuvers. Hence, in this paper, we present an integrated car-following and lane-changing decision-control system based on Deep Reinforcement Learning (DRL) to address this issue. Specifically, we consider a scenario where sudden construction work will be carried out along a highway. We model the scenario as a Markov Decision Process (MDP) and employ the well-known DQN algorithm to train the RL agent to make the appropriate decision accordingly (i.e., either stay in the same lane or change lanes). To overcome the delay and computational requirement of DRL algorithms, we adopt an MEC-assisted architecture where the RL agents are trained on MEC servers. We utilize the highly reputable SUMO simulator and OPENAI GYM to evaluate the performance of the proposed model under two policies; E-greedy policy and Boltzmann policy. The results unequivocally demonstrate that the DQN agent trained using the e-greedy policy significantly outperforms the one trained with the Boltzmann policy.

This paper describes the design and implementation of a real-time robotics system for semi-specular/painted surface defect detection. The system can be used on moving parts, tolerate varying lighting conditions, and can accommodate small inherent vibrations of the inspected surface that is common in manufacturing operations. Topographical information of the inspected surface is first obtained by the analysis of reflections of a known pattern from this surface. Spectral analysis is then applied to identify defects through novelty detection. Finally, a defect tracking mechanism eliminates spurious defects. The proposed system operates continuously at 90 fps. The paper presents field testing results that show the system can be used as a consistent and cost-effective way of quality control.

Transportation systems operate in a domain that is anything but simple. Many exhibit both spatial and temporal characteristics, at varying scales, under varying conditions brought on by external sources such as social events, holidays, and the weather. Yet, modeling the interplay of factors, devising generalized representations, and subsequently using them to solve a particular problem can be a challenging task. These situations represent only a fraction of the difficulties faced by modern intelligent transportation systems (ITS). In this paper, we present a survey that highlights the role modeling techniques within the realm of deep learning have played within ITS. We focus on how practitioners have formulated problems to address these various challenges, and outline both architectural and problem-specific considerations used to develop solutions. We hope this survey can help to serve as a bridge between the machine learning and transportation communities, shedding light on new domains and considerations in the future.

This paper presents a novel control strategy for a grasping free form unknown objects. The strategy adapts online the grasping forces applied at each contact point based on a bio-inspired criterion that reflects the total change in the measured force at each contact point. Experimental results with a two-fingered gripper equipped with tactile array sensors are provided. It shows that the adaptation can reach a stable grasp in less than 50ms from the initial contact.

This paper presents an integrated system that combines learning, a natural-language interface, and robotic grasping to enable the transfer of grasping skills from nontechnical users to robots. The system consists of two parts: a natural-language interface for grasping commands and a learning system. This paper focuses on the learning system and testing of the entire system in a small usability study. The learning system presented consists of two phases. In the first phase, the system learns to predict the next command, which the user is planning to issue based on command sequences recorded during previous grasping sessions. In the second phase, the system predicts the user's current state and moves the robot's gripper to the intended target endpoint to attempt to grasp the object. Using eight nontechnical users and a 5-degree-of-freedom (DOF) robot arm, a usability study was conducted to observe the impact of the learning system on user performance and satisfaction during a grasping operation. Experimental results show that the system was effective in learning users' grasping intentions, which allowed it to reduce the average time to grasp an object. In addition, participants' feedback from the usability study was generally positive toward having an adaptive robotics system that learns from their commands.

In this paper, we report on the findings of a human-robot interaction study that aims at developing a communication language for transferring grasping skills from a nontechnical user to a robot. Participants with different backgrounds and education levels were asked to command a five-degree-of-freedom human-scale robot arm to grasp five small everyday objects. They were allowed to use either commands from an existing command set or develop their own equivalent natural language instructions. The study revealed several important findings. First, individual participants were more inclined to use simple, familiar commands than more powerful ones. In most cases, once a set of instructions was found to accomplish the grasping task, few participants deviated from that set. In addition, we also found that the participant's background does appear to play a role during the interaction process. Overall, participants with less technical backgrounds require more time and more commands on average to complete a grasping task as compared to participants with more technical backgrounds.

In this paper, arithmetic representations for implementing multilayer perceptrons trained using the error backpropagation algorithm (MLP-BP) neural networks on field-programmable gate arrays (FPGAs) are examined in detail. Both floating-point (FLP) and fixed-point (FXP) formats are studied and the effect of precision of representation and FPGA area requirements are considered. A generic very high-speed integrated circuit hardware description language (VHDL) program was developed to help experiment with a large number of formats and designs. The results show that an MLP-BP network uses less clock cycles and consumes less real estate when compiled in an FXP format, compared with a larger and slower functioning compilation in an FLP format with similar data representation width, in bits, or a similar precision and range.

In this paper, we investigate the impact of the user's background on their ability to communicate grasping commands to a robot. We conducted a study where a group of 15 non-technical users use natural language to instruct a robotic arm to grasp five small everyday objects. We found that users with less technical backgrounds choose simple more predictable commands over complex unpredictable movements. These users also required more time and commands to complete a grasping task compared to users with more technical backgrounds. Other results however suggest that the user's background is not the most critical factor. Individual preferences and learning approaches also appear to play a role in command choices.

In this paper, a pilot study is conducted to explore developing a human-robot interaction language that specifically targets robotic grasping. The short term goal is to help nontechnical users command and control a simple robotic arm to grasp small objects. The long term objective is to use this language to enable skill transfer of grasping skills between nontechnical users and personal service robots. The study included a small group of participants with various technical backgrounds. They were asked to use a primitive set of commands to instruct a CRS robotic arm to grasp five small objects which are typically difficult to grasp. The findings of this pilot study are presented along with further insight gathered from participant feedback.

This paper present an architecture for combining a mixture of experts. The architecture has two unique features: 1) it assumes no prior knowledge of the size or structure of the mixture and allows the number of experts to dynamically expand during training, and 2) reinforcement feedback is used to guide the combining/expansion operation. The architecture is particularly suitable for applications when there is a need to approximate a many-to-many mapping. An example of such a problem is the task of training a robot to grasp arbitrarily shaped objects. This task requires the approximation of a many-to-many mapping, since various configurations can be used to grasp an object, and several objects can share the same grasping configuration. Experiments in a simulated environment using a 28-object database showed how the algorithm dynamically combined and expanded a mixture of neural networks to achieve the learning task. The paper also presents a comparison with two other nonlearning approaches.

The process of CAD model acquisition from existing objects is desirable in many industrial applications. An automated range image segmentation module is an essential building block and is key to speeding up this process. This paper presents a segmentation method based on local approximation of scan lines and uses adequate edge models to detect noise pixels as well as position and orientation discontinuities. This is followed by an adaptive grouping process to find a geometric representation of the different surface regions of the object. The output of the segmentation module is then used to automatically generate a surface CAD model of the scene. Experimental results on a large number of real range images demonstrate the efficiency and robustness of the method.

Presents a procedure for modeling and eliminating an end effector configuration error as a result of a faulty joint or a damaged link. This procedure provides an inexpensive software alternative to hardware replacement. A neural network model was developed and tested an a 6 DOF PUMA robot. The network approximates the error based on data obtained through observing the robot while executing a set of MOVE commands. The results show that, regardless of the error source, the robot's accuracy could be highly improved even when a small number of data points are used.

An experimental approach to robotic grasping is presented. This approach is based on developing a generic representation of grasping rules, which allows learning them from experiments between the object and the robot. A modular connectionist design arranged in subsumption layers is used to provide a mapping between sensory inputs and robot actions. Reinforcement feedback is used to select between different grasping rules and to reduce the number of failed experiments. This is particularly critical for applications in the personal service robot environment. Simulated experiments on a 15-object database show that the system is capable of learning grasping rules for each object in a finite number of experiments as well as generalizing from experiments on one object to grasping from another.

In this paper we present an experimental approach to robotic grasping that is based on mapping grasping rules to a generic representation that can then be learned by experiments. Furthermore, grasping rules acquired in this format can then be used on different objects using different grippers. During experimentation, reinforcement learning is used to minimize the number of failed experiments. Results show that the system is able to learn how to grasp various objects while maintaining a small number of experiments.