A Versatile Approach for Adaptive Grid Mapping and Grid Flex-Graph Exploration with a Field-Programmable Gate Array-Based Robot Using Hardware Schemes.

Robotic exploration in dynamic and complex environments requires advanced adaptive mapping strategies to ensure accurate representation of the environments. This paper introduces an innovative grid flex-graph exploration (GFGE) algorithm designed for single-robot mapping. This hardware-scheme-based algorithm leverages a combination of quad-grid and graph structures to enhance the efficiency of both local and global mapping implemented on a field-programmable gate array (FPGA). This novel research work involved using sensor fusion to analyze a robot's behavior and flexibility in the presence of static and dynamic objects. A behavior-based grid construction algorithm was proposed for the construction of a quad-grid that represents the occupancy of frontier cells. The selection of the next exploration target in a graph-like structure was proposed using partial reconfiguration-based frontier-graph exploration approaches. The complete exploration method handles the data when updating the local map to optimize the redundant exploration of previously explored nodes. Together, the exploration handles the quadtree-like structure efficiently under dynamic and uncertain conditions with a parallel processing architecture. Integrating several algorithms into indoor robotics was a complex process, and a Xilinx-based partial reconfiguration approach was used to prevent computing difficulties when running many algorithms simultaneously. These algorithms were developed, simulated, and synthesized using the Verilog hardware description language on Zynq SoC. Experiments were carried out utilizing a robot based on a field-programmable gate array (FPGA), and the resource utilization and power consumption of the device were analyzed.

Hardware Schemes for Smarter Indoor Robotics to Prevent the Backing Crash Framework Using Field Programmable Gate Array-Based Multi-Robots.

The use of smart indoor robotics services is gradually increasing in real-time scenarios. This paper presents a versatile approach to multi-robot backing crash prevention in indoor environments, using hardware schemes to achieve greater competence. Here, sensor fusion was initially used to analyze the state of multi-robots and their orientation within a static or dynamic scenario. The proposed novel hardware scheme-based framework integrates both static and dynamic scenarios for the execution of backing crash prevention. A round-robin (RR) scheduling algorithm was composed for the static scenario. Dynamic backing crash prevention was deployed by embedding a first come, first served (FCFS) scheduling algorithm. The behavioral control mechanism of the distributed multi-robots was integrated with FCFS and adaptive cruise control (ACC) scheduling algorithms. The integration of multiple algorithms is a challenging task for smarter indoor robotics, and the Xilinx-based partial reconfiguration method was deployed to avoid computational issues with multiple algorithms during the run-time. These methods were coded with Verilog HDL and validated using an FPGA (Zynq)-based multi-robot system.

A Versatile Approach to Polygonal Object Avoidance in Indoor Environments with Hardware Schemes Using an FPGA-Based Multi-Robot.

Service robots perform versatile functions in indoor environments. This study focuses on obstacle avoidance using flock-type indoor-based multi-robots. Each robot was developed with rendezvous behavior and distributed intelligence to perform obstacle avoidance. The hardware scheme-based obstacle-avoidance algorithm was developed using a bio-inspired flock approach, which was developed with three stages. Initially, the algorithm estimates polygonal obstacles and their orientations. The second stage involves performing avoidance at different orientations of obstacles using a heuristic based Bug2 algorithm. The final stage involves performing a flock rendezvous with distributed approaches and linear movements using a behavioral control mechanism. VLSI architectures were developed for multi-robot obstacle avoidance algorithms and were coded using Verilog HDL. The novel design of this article integrates the multi-robot's obstacle approaches with behavioral control and hardware scheme-based partial reconfiguration (PR) flow. The experiments were validated using FPGA-based multi-robots.

Hardware-Efficient Scheme for Trailer Robot Parking by Truck Robot in an Indoor Environment with Rendezvous.

Autonomous grounded vehicle-based social assistance/service robot parking in an indoor environment is an exciting challenge in urban cities. There are few efficient methods for parking multi-robot/agent teams in an unknown indoor environment. The primary objective of autonomous multi-robot/agent teams is to establish synchronization between them and to stay in behavioral control when static and when in motion. In this regard, the proposed hardware-efficient algorithm addresses the parking of a trailer (follower) robot in indoor environments by a truck (leader) robot with a rendezvous approach. In the process of parking, initial rendezvous behavioral control between the truck and trailer robots is established. Next, the parking space in the environment is estimated by the truck robot, and the trailer robot parks under the supervision of the truck robot. The proposed behavioral control mechanisms were executed between heterogenous-type computational-based robots. Optimized sensors were used for traversing and the execution of the parking methods. The truck robot leads, and the trailer robot mimics the actions in the execution of path planning and parking. The truck robot was integrated with FPGA (Xilinx Zynq XC7Z020-CLG484-1), and the trailer was integrated with Arduino UNO computing devices; this heterogenous modeling is adequate in the execution of trailer parking by a truck. The hardware schemes were developed using Verilog HDL for the FPGA (truck)-based robot and Python for the Arduino (trailer)-based robot.

Pareto optimization with small data by learning across common objective spaces.

In multi-objective optimization, it becomes prohibitively difficult to cover the Pareto front (PF) as the number of points scales exponentially with the dimensionality of the objective space. The challenge is exacerbated in expensive optimization domains where evaluation data is at a premium. To overcome insufficient representations of PFs, Pareto estimation (PE) invokes inverse machine learning to map preferred but unexplored regions along the front to the Pareto set in decision space. However, the accuracy of the inverse model depends on the training data, which is inherently scarce/small given high-dimensional/expensive objectives. To alleviate this small data challenge, this paper marks a first study on multi-source inverse transfer learning for PE. A method to maximally utilize experiential source tasks to augment PE in the target optimization task is proposed. Information transfers between heterogeneous source-target pairs is uniquely enabled in the inverse setting through the unification provided by common objective spaces. Our approach is tested experimentally on benchmark functions as well as on high-fidelity, multidisciplinary simulation data of composite materials manufacturing processes, revealing significant gains to the predictive accuracy and PF approximation capacity of Pareto set learning. With such accurate inverse models made feasible, a future of on-demand human-machine interaction facilitating multi-objective decisions is envisioned.

ACSL: Adaptive correlation-driven sparsity learning for deep neural network compression.

Deep convolutional neural network compression has attracted lots of attention due to the need to deploy accurate models on resource-constrained edge devices. Existing techniques mostly focus on compressing networks for image-level classification, and it is not clear if they generalize well on network architectures for more challenging pixel-level tasks, e.g., dense crowd counting or semantic segmentation. In this paper, we propose an adaptive correlation-driven sparsity learning (ACSL) framework for channel pruning that outperforms state-of-the-art methods on both image-level and pixel-level tasks. In our ACSL framework, we first quantify the data-dependent channel correlation information with a channel affinity matrix. Next, we leverage these inter-dependencies to induce sparsity into the channels with the introduced adaptive penalty strength. After removing the redundant channels, we obtain compact and efficient models, which have significantly less number of parameters while maintaining comparable performance with the original models. We demonstrate the advantages of our proposed approach on three popular vision tasks, i.e., dense crowd counting, semantic segmentation, and image-level classification. The experimental results demonstrate the superiority of our framework. In particular, for crowd counting on the Mall dataset, the proposed ACSL framework is able to reduce up to 94% parameters (VGG16-Decoder) and 84% FLOPs (ResNet101), while maintaining the same performance of (at times outperforming) the original model.