Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432806
Siyuan Li, Jingwen Wei
This paper considers a scenario in which a solar-powered UAV travels in a dynamic urban environment with unknown static and moving obstacles. A framework is proposed, including a three-phase hybrid approach for this problem. Firstly, an energy-aware path planning algorithm is proposed based on the limited information of the environment. Secondly, a pure pursuit controller is applied to follow the pre-generated online path. Finally, an energy-aware reactive obstacle avoidance algorithm is used to avoid collision with unknown obstacles. The effectiveness of the proposed framework is verified based on computer simulation.
{"title":"A Hybrid Approach for Navigation of a Solar-powered UAV in a Dynamic Urban Environment","authors":"Siyuan Li, Jingwen Wei","doi":"10.1109/ANZCC59813.2024.10432806","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432806","url":null,"abstract":"This paper considers a scenario in which a solar-powered UAV travels in a dynamic urban environment with unknown static and moving obstacles. A framework is proposed, including a three-phase hybrid approach for this problem. Firstly, an energy-aware path planning algorithm is proposed based on the limited information of the environment. Secondly, a pure pursuit controller is applied to follow the pre-generated online path. Finally, an energy-aware reactive obstacle avoidance algorithm is used to avoid collision with unknown obstacles. The effectiveness of the proposed framework is verified based on computer simulation.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"39 3","pages":"137-142"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432923
Dao Zhou, Yunlong Yang, Xiongjun Wu, Biao Liu
In conventional servo system with transmission mechanisms, the transmission backlash seems to be an inevitable phenomenon in many applications. Excessive backlash can lead to relative motion between the servo motor and the load, resulting in response delays, instability, and oscillations, consequently causing a decline in the dynamic performance of the system. To mitigate the negative effects of transmission backlash, this paper proposes a model-based backstepping adaptive control algorithm and a switching compensation approach tailored to transmission backlash. A feed-forward control gap compensation method is suggested. Leveraging identified servo system parameter information, this method achieves compensation and suppression of the nonlinearity caused by backlash. The efficacy of the proposed method is verified through theoretical analysis and simulation experiments, which also guaranteed its capability to effectively enhance the tracking accuracy and stability of the servo system while suppressing undesirable phenomena caused by backlash.
{"title":"Suppression of Adverse Effects of Transmission Clearance in Brushless DC Motor Servo Systems by Switching Compensation*","authors":"Dao Zhou, Yunlong Yang, Xiongjun Wu, Biao Liu","doi":"10.1109/ANZCC59813.2024.10432923","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432923","url":null,"abstract":"In conventional servo system with transmission mechanisms, the transmission backlash seems to be an inevitable phenomenon in many applications. Excessive backlash can lead to relative motion between the servo motor and the load, resulting in response delays, instability, and oscillations, consequently causing a decline in the dynamic performance of the system. To mitigate the negative effects of transmission backlash, this paper proposes a model-based backstepping adaptive control algorithm and a switching compensation approach tailored to transmission backlash. A feed-forward control gap compensation method is suggested. Leveraging identified servo system parameter information, this method achieves compensation and suppression of the nonlinearity caused by backlash. The efficacy of the proposed method is verified through theoretical analysis and simulation experiments, which also guaranteed its capability to effectively enhance the tracking accuracy and stability of the servo system while suppressing undesirable phenomena caused by backlash.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"1056 1","pages":"131-136"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article presents an innovative modular planting plant factory inspired by lean manufacturing principle, where planting cells work in a distributed way, coordinated by cloud service. Each planting cell is equipped with controllable light source, CO2 supplement and fans. All cells are remotely and locally controllable, including instructions on environmental adjustment and planting task scheduling. This system offers flexibility, scalability, and autonomous management of plant factories. Preliminary results indicate its potential in precision agriculture. The modular system adapts to diverse scenarios, focusing on robustness in smart factories and user-friendliness. It envisions urban cultivation as a social platform and aligns with emerging trends in cyber-physical social systems (CPSS) and Distributed Autonomous Organization (DAO). This work represents a convergence of technology, adaptability, and a vision for interactive urban agriculture.
{"title":"Easy Planting with Distributed Cellular Plant Factory*","authors":"Jing Hua, Xiujuan Wang, Haoyu Wang, Xingyuan Dai, Weikang Zhao, Mengzhen Kang","doi":"10.1109/ANZCC59813.2024.10432816","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432816","url":null,"abstract":"This article presents an innovative modular planting plant factory inspired by lean manufacturing principle, where planting cells work in a distributed way, coordinated by cloud service. Each planting cell is equipped with controllable light source, CO2 supplement and fans. All cells are remotely and locally controllable, including instructions on environmental adjustment and planting task scheduling. This system offers flexibility, scalability, and autonomous management of plant factories. Preliminary results indicate its potential in precision agriculture. The modular system adapts to diverse scenarios, focusing on robustness in smart factories and user-friendliness. It envisions urban cultivation as a social platform and aligns with emerging trends in cyber-physical social systems (CPSS) and Distributed Autonomous Organization (DAO). This work represents a convergence of technology, adaptability, and a vision for interactive urban agriculture.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"73 ","pages":"207-210"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432796
A. Savkin, S. Verma, Siyuan Li
This paper presents a solution to the complex challenge of optimizing the navigation of unmanned mining trucks across challenging terrain. The primary goal is to minimize fuel consumption while ensuring safety by avoiding designated hazard zones. The developed algorithm efficiently calculates an optimal path for the truck by considering both the terrain’s geometric characteristics and safety constraints. Importantly, this research rigorously proves the global optimality of the proposed navigation algorithm. Our work contributes to enhancing the efficiency and safety of autonomous mining operations.
{"title":"Energy-Efficient Off-Road Navigation of an Unmanned Mining Truck on a Rough Terrain","authors":"A. Savkin, S. Verma, Siyuan Li","doi":"10.1109/ANZCC59813.2024.10432796","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432796","url":null,"abstract":"This paper presents a solution to the complex challenge of optimizing the navigation of unmanned mining trucks across challenging terrain. The primary goal is to minimize fuel consumption while ensuring safety by avoiding designated hazard zones. The developed algorithm efficiently calculates an optimal path for the truck by considering both the terrain’s geometric characteristics and safety constraints. Importantly, this research rigorously proves the global optimality of the proposed navigation algorithm. Our work contributes to enhancing the efficiency and safety of autonomous mining operations.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"1217 34","pages":"98-102"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432847
Roxanne R. Jackson, Damiano Varagnolo, S. Knorn
Biofeedback in gamified medical exercise sessions has proven to be an effective technique for adapting patient behaviours to improve health outcomes. In this paper, we formulate a method for designing optimal training sessions with two conflicting goals: maximising the desired exercise effect and sufficiently exciting the system for identification in order to update the personalised patient model. We exploit the flexibility of model-free reinforcement learning to obtain an optimal controller, which is robust to uncertainty in the system parameters. We compare the controller from reinforcement learning to a standard dual control formulation in simulation on an illustrative case study of building pelvic floor muscular strength and tone while performing Kegel exercises. The results indicate that the reinforcement learning method attains a better exercise effect while improving the parameter estimates compared to a standard dual controller. However, due to the trial-and-error nature of reinforcement learning, this comes at the expense of computational time.
{"title":"Reinforcement Learning for Exploration vs. Exploitation Problems in Medical Exercise Sessions","authors":"Roxanne R. Jackson, Damiano Varagnolo, S. Knorn","doi":"10.1109/ANZCC59813.2024.10432847","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432847","url":null,"abstract":"Biofeedback in gamified medical exercise sessions has proven to be an effective technique for adapting patient behaviours to improve health outcomes. In this paper, we formulate a method for designing optimal training sessions with two conflicting goals: maximising the desired exercise effect and sufficiently exciting the system for identification in order to update the personalised patient model. We exploit the flexibility of model-free reinforcement learning to obtain an optimal controller, which is robust to uncertainty in the system parameters. We compare the controller from reinforcement learning to a standard dual control formulation in simulation on an illustrative case study of building pelvic floor muscular strength and tone while performing Kegel exercises. The results indicate that the reinforcement learning method attains a better exercise effect while improving the parameter estimates compared to a standard dual controller. However, due to the trial-and-error nature of reinforcement learning, this comes at the expense of computational time.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"385 ","pages":"55-60"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Gun-launched missiles play a significant role in land battlefields. However, simulating its physical properties realistically within training environments remains a persistent challenge, particularly when it comes to replicating the dynamic characteristics between virtual and real interactions. To address this challenge, we propose an extensive framework of theoretical methods aimed at achieving precision strikes on arbitrary moving targets in three-dimensional space based on the three-point method. Firstly, we establish the external ballistic kinematics and dynamics equations of the gun-launched missile. Then, we conduct the overall structural design and three-dimensional modeling of the missile, taking the ‘9K120’ gun-launched missile as the reference model. Next, we compute the aerodynamic coefficients and the aerodynamic moment coefficients of the gun-launched missile. Finally, we establish a simulation model of the gun-launched missile and analyze the curves of the external ballistic trajectory and flight attitude change over time when the missile strikes the target, and we finish mathematical modeling of gun-launched missile ballistic trajectories in LVC architecture. The results demonstrate that our theoretical method effectively solves the simulation modeling problem of the gun-launched missile in the simulation training environment, enabling accurate strikes on moving targets in three-dimensional space and achieving interactive shooting training.
{"title":"Achieving Interactive Shooting Training: Mathematical Modeling of Gun-launched Missile Ballistic Trajectories in LVC Architecture","authors":"Dongqing Li, Yin Zhu, Yuanwu Zhu, Xinyu Wu, Hongmei Zhang, Yuan Liu","doi":"10.1109/ANZCC59813.2024.10432825","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432825","url":null,"abstract":"Gun-launched missiles play a significant role in land battlefields. However, simulating its physical properties realistically within training environments remains a persistent challenge, particularly when it comes to replicating the dynamic characteristics between virtual and real interactions. To address this challenge, we propose an extensive framework of theoretical methods aimed at achieving precision strikes on arbitrary moving targets in three-dimensional space based on the three-point method. Firstly, we establish the external ballistic kinematics and dynamics equations of the gun-launched missile. Then, we conduct the overall structural design and three-dimensional modeling of the missile, taking the ‘9K120’ gun-launched missile as the reference model. Next, we compute the aerodynamic coefficients and the aerodynamic moment coefficients of the gun-launched missile. Finally, we establish a simulation model of the gun-launched missile and analyze the curves of the external ballistic trajectory and flight attitude change over time when the missile strikes the target, and we finish mathematical modeling of gun-launched missile ballistic trajectories in LVC architecture. The results demonstrate that our theoretical method effectively solves the simulation modeling problem of the gun-launched missile in the simulation training environment, enabling accurate strikes on moving targets in three-dimensional space and achieving interactive shooting training.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"29 ","pages":"43-48"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432921
Moslem Uddin, Huadong Mo, Daoyi Dong, S. Elsawah
This study proposes an energy management framework for community microgrids to produce energy at a reduced cost with high reliability. The techno-economic benefits of the proposed solution are also investigated. A regional community with solar and wind energies is considered as a case study. The test microgrid consists of solar photovoltaic, wind turbines, a diesel generator, battery storage, and grid connection capability. Simple performance metrics are used to assess the performance of the proposed energy management framework. Finally, results are compared with and without energy management implementation. Results show that the proposed framework reduces electricity costs by up to 18.50% compared to a baseline method. Reliability analysis reveals that the proposed methods ensure continuous power supply even during grid outages. This study also finds that the proposed energy management technique reduces fuel consumption by approximately 90.25%. This work can be valuable in identifying effective solutions for supplying reliable and cost-effective power to regional and remote areas.
{"title":"Energy Management in Community Microgrid—A Techno-economic Study Framework","authors":"Moslem Uddin, Huadong Mo, Daoyi Dong, S. Elsawah","doi":"10.1109/ANZCC59813.2024.10432921","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432921","url":null,"abstract":"This study proposes an energy management framework for community microgrids to produce energy at a reduced cost with high reliability. The techno-economic benefits of the proposed solution are also investigated. A regional community with solar and wind energies is considered as a case study. The test microgrid consists of solar photovoltaic, wind turbines, a diesel generator, battery storage, and grid connection capability. Simple performance metrics are used to assess the performance of the proposed energy management framework. Finally, results are compared with and without energy management implementation. Results show that the proposed framework reduces electricity costs by up to 18.50% compared to a baseline method. Reliability analysis reveals that the proposed methods ensure continuous power supply even during grid outages. This study also finds that the proposed energy management technique reduces fuel consumption by approximately 90.25%. This work can be valuable in identifying effective solutions for supplying reliable and cost-effective power to regional and remote areas.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"589 ","pages":"169-174"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432831
Omar Zakary, Ke-Xian Liu, Chao Ren, Qing-Hao Meng
The multi-joint structure of our Autonomous Underwater Vehicle (AUV) enhances its maneuverability, allowing it to navigate in the underwater environment with greater flexibility. However, this added maneuverability poses challenges to the steering process. When the multi-joint AUV (MJ-AUV) performs steering maneuvers, its joints undergo rotation, leading to a change in the orientation of the cabins with respect to the overall forward heading of the vehicle. As a result, this change in orientation affects the values of the hydrodynamic reactions, including Coriolis and damping, experienced by the cabins. The dynamic interaction between the joints' rotation and the resulting change in hydrodynamic forces significantly impacts the steering performance and stability of the MJ-AUV. Understanding and addressing these effects are crucial for the development of effective control strategies that ensure precise and reliable steering in various underwater environments. Overcoming the difficulties posed by joint rotation during steering can lead to advancements in MJ-AUV navigation and expand their potential applications in complex underwater missions. This research proposes a fuzzy-based control with sliding mode control (SMC) approach for the steering of an MJAUV. The developed fuzzy-based SMC control algorithm is validated through extensive MATLAB simulations. The results demonstrate improved tracking performance and robustness in comparison to the SMC control method. Moreover, the proposed approach shows superior trajectory tracking accuracy while mitigating undesired chattering effects associated with standard SMC techniques.
{"title":"Fuzzy Based Steering Control of a Multi-Joint AUV","authors":"Omar Zakary, Ke-Xian Liu, Chao Ren, Qing-Hao Meng","doi":"10.1109/ANZCC59813.2024.10432831","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432831","url":null,"abstract":"The multi-joint structure of our Autonomous Underwater Vehicle (AUV) enhances its maneuverability, allowing it to navigate in the underwater environment with greater flexibility. However, this added maneuverability poses challenges to the steering process. When the multi-joint AUV (MJ-AUV) performs steering maneuvers, its joints undergo rotation, leading to a change in the orientation of the cabins with respect to the overall forward heading of the vehicle. As a result, this change in orientation affects the values of the hydrodynamic reactions, including Coriolis and damping, experienced by the cabins. The dynamic interaction between the joints' rotation and the resulting change in hydrodynamic forces significantly impacts the steering performance and stability of the MJ-AUV. Understanding and addressing these effects are crucial for the development of effective control strategies that ensure precise and reliable steering in various underwater environments. Overcoming the difficulties posed by joint rotation during steering can lead to advancements in MJ-AUV navigation and expand their potential applications in complex underwater missions. This research proposes a fuzzy-based control with sliding mode control (SMC) approach for the steering of an MJAUV. The developed fuzzy-based SMC control algorithm is validated through extensive MATLAB simulations. The results demonstrate improved tracking performance and robustness in comparison to the SMC control method. Moreover, the proposed approach shows superior trajectory tracking accuracy while mitigating undesired chattering effects associated with standard SMC techniques.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"85 ","pages":"72-77"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432849
Haifang Song, Derui Ding, Qing-Long Han, Xiaohua Ge
This article proposes a novel trust-based entropy filter in distributed form for state-saturated nonlinear systems with hybrid cyber-attacks, including denial-of-service and deception attacks. A simple clustering method is employed to categorize the data received from neighboring nodes into two clusters: the trusted cluster and the untrusted cluster. By optimizing a joint cost function involving weighted least squares and a generalized maximum correntropy criterion, a two-step filter is designed in a distributed form, wherein the untrusted data from the neighbors is compensated to relieve the impact from malicious attacks. The significance of the designed algorithm is verified by conducting a target-tracking experiment at the end.
{"title":"Trust-Based Distributed Entropy Filtering for State-Saturated Nonlinear Systems with Hybrid Cyber-Attacks and Non-Gaussian Noises","authors":"Haifang Song, Derui Ding, Qing-Long Han, Xiaohua Ge","doi":"10.1109/ANZCC59813.2024.10432849","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432849","url":null,"abstract":"This article proposes a novel trust-based entropy filter in distributed form for state-saturated nonlinear systems with hybrid cyber-attacks, including denial-of-service and deception attacks. A simple clustering method is employed to categorize the data received from neighboring nodes into two clusters: the trusted cluster and the untrusted cluster. By optimizing a joint cost function involving weighted least squares and a generalized maximum correntropy criterion, a two-step filter is designed in a distributed form, wherein the untrusted data from the neighbors is compensated to relieve the impact from malicious attacks. The significance of the designed algorithm is verified by conducting a target-tracking experiment at the end.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"232 3","pages":"164-168"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.1109/ANZCC59813.2024.10432872
Fenghua Zhu, Yuanyuan Chen
Lane line detection is an important input of the automatic driving system and the assisted driving system. It is deployed on the vehicle end, with limited resources and high requirements for real-time performance and detection accuracy. We explore a new knowledge distillation method for lane line detection, in which the student network can acquire knowledge not only from the output features of the teacher network but also from the intermediate process of the teacher network. The knowledge distillation in the intermediate process named important feature correlations distillation compares the correlation between the feature maps of the teacher network and the student network. The knowledge distillation of the output results named semantic consistency distillation allows the student network to learn the output feature knowledge of the teacher network by integrating confrontation training into the knowledge distillation method. Experimental results demonstrate that our knowledge distillation method works well and light models can benefit from the distillation method.
{"title":"A Knowledge Distillation Network Combining Adversarial Training and Intermediate Feature Extraction for Lane Line Detection","authors":"Fenghua Zhu, Yuanyuan Chen","doi":"10.1109/ANZCC59813.2024.10432872","DOIUrl":"https://doi.org/10.1109/ANZCC59813.2024.10432872","url":null,"abstract":"Lane line detection is an important input of the automatic driving system and the assisted driving system. It is deployed on the vehicle end, with limited resources and high requirements for real-time performance and detection accuracy. We explore a new knowledge distillation method for lane line detection, in which the student network can acquire knowledge not only from the output features of the teacher network but also from the intermediate process of the teacher network. The knowledge distillation in the intermediate process named important feature correlations distillation compares the correlation between the feature maps of the teacher network and the student network. The knowledge distillation of the output results named semantic consistency distillation allows the student network to learn the output feature knowledge of the teacher network by integrating confrontation training into the knowledge distillation method. Experimental results demonstrate that our knowledge distillation method works well and light models can benefit from the distillation method.","PeriodicalId":518506,"journal":{"name":"2024 Australian & New Zealand Control Conference (ANZCC)","volume":"18 9","pages":"92-97"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140528541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：