In recent years, the rapid advancement of robot technology has transformed various industries, leading to an increased demand for engineers who can tackle complex engineering problems. As educators, we recognize the importance of integrating practical experiences into the curriculum to prepare students for real-world challenges. Complex engineering problems, characterized by their multifaceted nature, require a holistic approach that combines theoretical knowledge with hands-on application. In our institution, we have developed a comprehensive practical teaching system centered on intelligent robots, which serve as an ideal platform for embodying these complexities. This system emphasizes two key directions: intelligent perception and motion control, both critical components of modern robot technology. By aligning our teaching methods with the requirements of engineering education accreditation, we aim to cultivate students’ abilities to analyze, design, and solve intricate issues in robot technology.
The current state of practical teaching in robot technology often faces challenges such as fragmentation between courses and a lack of systemic integration. Many programs offer isolated experiments that do not fully capture the interconnectedness of disciplines like control theory, sensor integration, and artificial intelligence. This can lead to a superficial understanding, where students accumulate knowledge without the ability to apply it cohesively. In our experience, this disconnect hinders the development of skills needed to address complex engineering problems in robot technology. For instance, while students may learn about PID control in one course and sensor fusion in another, they rarely get the opportunity to combine these elements in a unified project. Therefore, we have restructured our practical teaching system to bridge these gaps, ensuring that each component builds upon the previous one and fosters a deeper comprehension of robot technology.
Our overall approach to constructing this practical teaching system revolves around a hierarchical and progressive framework. We begin by identifying the core aspects of robot technology that align with complex engineering problems, such as uncertainty in sensor data, non-linear dynamics in motion control, and the integration of multiple subsystems. The intelligent perception direction focuses on enabling robots to interpret their environment using sensors like cameras, lidar, and inertial measurement units, while the motion control direction deals with algorithms for navigation, path planning, and actuator management. This dual focus allows students to explore the full spectrum of robot technology, from data acquisition to decision-making. By gradually increasing the complexity of tasks, we help students develop the ability to handle open-ended problems that lack straightforward solutions, a hallmark of complex engineering issues in robot technology.
To integrate robot technology into the foundational practices of our curriculum, we have mapped key courses to specific functional modules of intelligent robots. For example, courses like “Sensor Application Technology” and “Automatic Control Principles” are linked to experiments involving environmental perception and feedback systems. The table below illustrates how various robot devices correspond to different courses, providing a structured pathway for students to connect theoretical concepts with practical applications in robot technology.
| Device Name | Supported Courses | Experiment Hours |
|---|---|---|
| AI Smart Car with Deep Learning | Robotics Fundamentals, Artificial Intelligence and Machine Learning | 8, 10 |
| Tank Robot with Robotic Arm | Automatic Detection Technology, STM32 Principles and Applications | 6, 10 |
| Unmanned Smart Vehicle Model | Robotics Fundamentals, Artificial Intelligence and Machine Learning | 8, 10 |
| Ship Intelligent Perception and Control Platform | Automatic Control Principles, Data Acquisition and Processing Technology | 12, 8 |
| Underwater Robot with Manipulator | Automatic Control Principles, Integrated Design of Measurement and Control Systems | 12, 60 |
| Bionic Robotic Fish | Automatic Control Principles, Integrated Design of Measurement and Control Systems | 12, 60 |
| Quadrotor Flight Control System | Automatic Control Principles, Integrated Design of Measurement and Control Systems | 12, 60 |
In the early stages of their education, students engage with basic programming and assembly of robots, such as Arduino-based wheeled robots or RoboMaster S1 models. This introduces them to fundamental concepts in robot technology, including motor control and sensor interfacing. As they progress, they tackle more advanced topics like sensor fusion, where data from multiple sources must be combined to achieve accurate perception. For instance, in intelligent perception experiments, students work with infrared sensors and encoders to implement obstacle avoidance, which requires understanding probabilistic models and filtering techniques. A common formula used in this context is the Kalman filter, which can be represented as: $$ \hat{x}_{k|k} = \hat{x}_{k|k-1} + K_k(z_k – H_k\hat{x}_{k|k-1}) $$ where $\hat{x}_{k|k}$ is the updated state estimate, $K_k$ is the Kalman gain, $z_k$ is the measurement, and $H_k$ is the observation matrix. This equation helps students grasp how robot technology leverages mathematical tools to reduce uncertainty in complex environments.
Motion control in robot technology often involves non-linear systems that demand sophisticated algorithms. We introduce students to PID control initially, with the standard form: $$ u(t) = K_p e(t) + K_i \int_0^t e(\tau) d\tau + K_d \frac{de(t)}{dt} $$ where $u(t)$ is the control output, $e(t)$ is the error, and $K_p$, $K_i$, $K_d$ are the proportional, integral, and derivative gains, respectively. However, to address complex engineering problems, we extend this to adaptive control and machine learning methods. For example, neural networks can be employed for path planning, with the output given by: $$ y = f\left(\sum_{i=1}^n w_i x_i + b\right) $$ where $y$ is the decision output, $x_i$ are inputs such as sensor data, $w_i$ are weights, $b$ is the bias, and $f$ is an activation function. Through these mathematical representations, students learn to model and simulate behaviors in robot technology, enhancing their problem-solving skills for scenarios where traditional methods fall short.
Competitions play a pivotal role in enriching our practical teaching system for robot technology. Events like robotics tournaments challenge students to apply their knowledge in dynamic, unpredictable settings, mirroring real-world complex engineering problems. We have adapted competition mechanisms into our curriculum by simplifying rules and incorporating relevant modules into course designs. For instance, tasks such as autonomous navigation or object recognition in contests become the basis for comprehensive projects in courses like “Data Acquisition and Processing Technology.” This not only motivates students but also fosters innovation and teamwork, key aspects of advancing robot technology. By participating in these activities, students gain exposure to cutting-edge developments and learn to iterate on their designs, dealing with issues like sensor noise or actuator delays that are common in complex robot technology applications.
Our hardware and software resources form the backbone of this practical teaching system in robot technology. We have equipped laboratories with a diverse range of devices, including AI smart cars, robotic arms, and underwater robots, all supported by software tools like MATLAB, Python, and embedded programming environments. These resources enable students to experiment with various aspects of robot technology, from low-level control to high-level decision-making. For example, the deep learning AI smart car allows students to implement convolutional neural networks for image recognition, a crucial skill in intelligent perception. The integration of these tools ensures that students can tackle complex engineering problems from multiple angles, reinforcing the interdisciplinary nature of robot technology.
The implementation of our practical teaching system in robot technology is divided into three progressive stages, each designed to build upon the previous one. In the first stage, students focus on basic robot assembly and programming, such as controlling motors to follow predefined trajectories. This introduces them to fundamental concepts in robot technology and helps them develop a hands-on understanding of mechanical and electronic components. As they move to the second stage, they begin integrating sensors and designing more complex control logic, such as using encoders for precise positioning or implementing simple AI algorithms for decision-making. This phase emphasizes the integration of knowledge from various courses, addressing complex engineering problems like multi-sensor calibration and real-time data processing in robot technology.
In the third stage, students engage in advanced projects that require innovation and independent research. They might work on autonomous vehicles that use lidar and cameras for environment mapping, or underwater robots that perform tasks in uncertain conditions. These projects often involve optimizing algorithms for efficiency and robustness, such as using genetic algorithms for path planning: $$ F(x) = \sum_{i=1}^n w_i f_i(x) $$ where $F(x)$ is the fitness function, $f_i(x)$ are objective functions like path length or energy consumption, and $w_i$ are weights. Through these experiences, students learn to handle the unpredictability and multi-objective nature of complex engineering problems in robot technology, preparing them for careers in research and development. Overall, this staged approach ensures that students gradually develop the confidence and competence to tackle real-world challenges in robot technology.
In conclusion, our practical teaching system for intelligent robots has proven effective in enhancing students’ abilities to solve complex engineering problems. By focusing on intelligent perception and motion control, and leveraging competitions and integrated resources, we have created a dynamic learning environment that mirrors the demands of modern robot technology. The use of formulas and tables helps solidify theoretical concepts, while hands-on projects foster practical skills. As robot technology continues to evolve, we remain committed to refining this system, ensuring that our graduates are well-equipped to contribute to this exciting field. Through continuous iteration and feedback, we aim to set a benchmark for practical education in robot technology, empowering the next generation of engineers to innovate and excel.