Author Affiliations
1College of Materials Science and Engineering, Guangdong Research Center for Interfacial Engineering of Functional Materials, Institute of Microelectronics (IME), Shenzhen University, Shenzhen 518060, China2Department of Photonics & Institute of Electro-Optical Engineering, College of Electrical and Computer Engineering, Taiwan Yang Ming Chiao Tung University & Taiwan Chiao Tung University, Hsinchu 30010, China3Department of Computer Science, University of Liverpool, Liverpool, UK4College of Materials Science and Engineering, Guangdong Research Center for Interfacial Engineering of Functional Materials, Institute of Microelectronics (IME), Shenzhen University, Shenzhen 518060, China5Department of Photonics & Institute of Electro-Optical Engineering, College of Electrical and Computer Engineering, Taiwan Yang Ming Chiao Tung University & Taiwan Chiao Tung University, Hsinchu 30010, Chinashow less
Fig. 1. (a) Schematic diagram of micro-LED backlight module; (b) schematic diagram of LED with DBR structure; (c) highly-reflective surface substrate; (d) etching structure of the receiver.
Fig. 2. Workflow for optimizing micro-LED backlight module: (a) the process of deep reinforcement learning and (b) the process of the virtual-realistic experiment.
Fig. 3. (a) Workflow of environment control agent and schematic diagram of the virtual-realistic experiment; (b) principle of kernel1: Gaussian and Lambertian reflection; (c) principle of kernel2: BSDF properties.
Fig. 4. Workflow of DDQN network.
Fig. 5. Virtual-realistic experiment: (a) single light pattern analysis; (b) module pattern analysis.
Fig. 6. Result of reinforcement learning: (a) uniformity for every iteration with reward function1; (b) uniformity for every iteration with reward function2; (c) uniformity for every iteration with reward function3; (d) the best result by reward function1; (e) the best result by reward function2; (f) the best result by reward function3.
Fig. 7. Demonstration of the micro-LED backlight module.
Fig. 8. Influence of DBR structure: (a) light pattern with state Sb3′, (b) light pattern with state Sb3, and (c) uniformity of Sb3 and Sb3′.
Fig. 9. Schematic of resonant loss for the micro-LED backlight module.
Action No. | Action Definition | Variation Symbol |
---|
| Add distance | | | Reduce distance | | | Add width | | | Reduce width | | | Add spacing | | | Reduce spacing | | | Add thickness | | | Reduce thickness | | | Add DBR pairs | DBR | | Reduce DBR pairs | DBR |
|
Table 1. Definition List of Action
State No. | State Definition | Variation symbol |
---|
| Value of distance | | | Value of spacing | | | Value of DBR pairs | | | Value of thickness | | | Value of width | DBR |
|
Table 2. Definition List of State
No. | Parameters | Variation Symbol | Range |
---|
1 | Distance from receiver to LED | | 100–180 μm | 2 | LED width | | 10–100 μm | 3 | LED spacing | | 300–700 μm | 4 | Thickness of LED | | 5–55 μm | 5 | DBR-BSDF pairs | DBR | 4.5–9.5 pairs |
|
Table 3. Definition List for Range of Parameters
No. | Parameters | Value |
---|
1 | Bulk of GaN-based light-emitting diodes | Refraction | 2 | High-reflectivity white bottom surface of module | Gaussian type 5° with 97% | 3 | Receiver reflectivity | Lambertian with 35% | 4 | Index of DBR material (AIN/GaN) | | 5 | Thickness of DBR material (AIN/GaN) | |
|
Table 4. Results of Virtual-Realistic Experiment
Best State | Reward Formula | Parameters | Number of Iterations | Best Uniformity |
---|
| | , , , , | 289 | 83.97% | | | , , , , | 183 | 86.51% | | | , , , , | 249 | 90.32% |
|
Table 5. Best Uniformity for Different Reward Functions
| Number of Iterations | Time (h) | Optimal Result |
---|
Entire loop | 106,920 | 297 | Uniformity: 89.61% | Designing agent | 16,000 | 53.3 | Uniformity: 90.32% |
|
Table 6. Work Efficiency of Designing Agent