Individual #5’s action ratio, position states. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

RSF Components of the best five individuals. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Open loop simulation. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Average wind test fitness. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Internal process of a policy gradient block. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Training process of a DDPG individual. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

PbGA search phases to find the best individuals. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Previous usages of DRL in solving PDG problems. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Internal process of a critic gradient block. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Best Individuals from the mapping phase. by Larasmoyo Nugroho (18078260)

“…<div><p>The PbGA-DDPG algorithm, which uses a potential-based GA-optimized reward shaping function, is a versatiledeep reinforcement learning/DRLagent that can control a vehicle in a complex environment without prior knowledge. …”

Schematic diagram of weld surface defects. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Improved YOLOv7 network structure. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Renderings of data enhancements. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Number and size of marked defects. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Loss function curve. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Precision-Recall curve. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Comparison experiment results. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Ablation experiment results. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Deepwise separable convolution structure diagram. by Xiangqian Xu (17310895)

“…<div><p>The background of pipeline weld surface defect image is complex, and the defect size is small. Aiming at the small defect size in the weld image, which is easy to cause missed detection and false detection, a lightweight target detection algorithm based on improved YOLOv7 is proposed. …”

Comparison analysis of computation time. by Indhumathi S. (19173013)

“…Furthermore, the matching score for the test image is 0.975. The computation time for CBFD is 2.8 ms, which is at least 6.7% lower than that of other algorithms. …”