Purpose: Visual perception enables robots to perceive the environment. Visual data is processed using computer vision algorithms that are usually time-expensive and require powerful devices to process the visual data in real-time, which is unfeasible for open-field robots with limited energy. This work benchmarks the performance of different heterogeneous platforms for object detection in real-time. This research benchmarks three architectures: embedded GPU -- Graphical Processing Units (such as NVIDIA Jetson Nano 2 GB and 4 GB, and NVIDIA Jetson TX2), TPU -- Tensor Processing Unit (such as Coral Dev Board TPU), and DPU -- Deep Learning Processor Unit (such as in AMD-Xilinx ZCU104 Development Board, and AMD-Xilinx Kria KV260 Starter Kit). Method: The authors used the RetinaNet ResNet-50 fine-tuned using the natural VineSet dataset. After the trained model was converted and compiled for target-specific hardware formats to improve the execution efficiency. Conclusions and Results: The platforms were assessed in terms of performance of the evaluation metrics and efficiency (time of inference). Graphical Processing Units (GPUs) were the slowest devices, running at 3 FPS to 5 FPS, and Field Programmable Gate Arrays (FPGAs) were the fastest devices, running at 14 FPS to 25 FPS. The efficiency of the Tensor Processing Unit (TPU) is irrelevant and similar to NVIDIA Jetson TX2. TPU and GPU are the most power-efficient, consuming about 5W. The performance differences, in the evaluation metrics, across devices are irrelevant and have an F1 of about 70 % and mean Average Precision (mAP) of about 60 %.
翻译:目的: 视觉感知使机器人能够感知环境。 视觉数据是使用计算机视觉算法处理的, 计算机视觉算法通常耗时甚长, 需要强大的设备实时处理视觉数据, 对于能量有限的露天机器人来说, 这是不可行的。 这项工作基准了不同多元平台的性能, 实时检测物体。 方法 : 作者使用 内嵌 GPU -- 图形处理器( 如 NVIDIA Jetson Nano 2 GB 和 4GB 和 NVIDIA Jetson TX2 ), 以及 NVIDIA 1 Jetson TX2 、 TPU - 登光处理器处理器( 如 Coloral Developing TCU) 和 DPU- 深学习处理器( 如 AM- Xilinx ZCU104 开发委员会 和 AM XIKIK Kria KIR ) 。 方法: 使用 RetinNetNetNet- Reset the Reduction press press and FPI laft FPI press 和 FPIPIPI 5 快速处理器效率。 在FPI 5 5 和 FPIPI 标准中, 5 和FPIPIPS 运行中, 5 和FPI 5 和FPIPS 5 节节节中, 。 和FPIPS 的运行中, 。