Efficient trajectory generation in complex dynamic environments remains an open problem in the operation of an unmanned surface vehicle (USV). The perception of a USV is usually interfered by the swing of the hull and the ambient weather, making it challenging to plan optimal USV trajectories. In this paper, a cooperative trajectory planning algorithm for a coupled USV-UAV system is proposed to ensure that a USV can execute a safe and smooth path as it autonomously advances through multi-obstacle maps. Specifically, the unmanned aerial vehicle (UAV) plays the role of a flight sensor, providing real-time global map and obstacle information with a lightweight semantic segmentation network and 3D projection transformation. An initial obstacle avoidance trajectory is generated by a graph-based search method. Concerning the unique under-actuated kinematic characteristics of the USV, a numerical optimization method based on hull dynamic constraints is introduced to make the trajectory easier to be tracked for motion control. Finally, a motion control method based on NMPC with the lowest energy consumption constraint during execution is proposed. Experimental results verify the effectiveness of the whole system, and the generated trajectory is locally optimal for USV with considerable tracking accuracy.

Data simulation engines like Unity are becoming an increasingly important data source that allows us to acquire ground truth labels conveniently. Moreover, we can flexibly edit the content of an image in the engine, such as objects (position, orientation) and environments (illumination, occlusion). When using simulated data as training sets, its editable content can be leveraged to mimick the distribution of real-world data, and thus reduce the content difference between the synthetic and real domains. This paper explores content adaptation in the context of semantic segmentation, where the complex street scenes are fully synthesized using 19 classes of virtual objects from a first person driver perspective and controlled by 23 attributes. To optimize the attribute values and obtain a training set of similar content to real-world data, we propose a scalable discretization-and-relaxation (SDR) approach. Under a reinforcement learning framework, we formulate attribute optimization as a random-to-optimized mapping problem using a neural network. Our method has three characteristics. 1) Instead of editing attributes of individual objects, we focus on global attributes that have large influence on the scene structure, such as object density and illumination. 2) Attributes are quantized to discrete values, so as to reduce search space and training complexity. 3) Correlated attributes are jointly optimized in a group, so as to avoid meaningless scene structures and find better convergence points. Experiment shows our system can generate reasonable and useful scenes, from which we obtain promising real-world segmentation accuracy compared with existing synthetic training sets.

In this research, we present an intelligent forklift cargo precision transfer system to address the issue of poor pallet docking accuracy and low recognition rate when using current techniques. The technology is primarily used to automatically check if there is any pallet that need to be transported. The intelligent forklift is then sent to the area of the target pallet after being recognized. Images of the pallets are then collected using the forklift’s camera, and a deep learning-based recognition algorithm is used to calculate the precise position of the pallets. Finally, the forklift is controlled by a high-precision control algorithm to insert the pallet in the exact location. This system creatively introduces the small target detection into the pallet target recognition system, which greatly improves the recognition rate of the system. The application of Yolov5 into the pallet positional calculation makes the coverage and recognition accuracy of the algorithm improved. In comparison with the prior approach, this system’s identification rate and accuracy are substantially higher, and it requires fewer sensors and indications to help with deployment. We have collected a significant amount of real data in order to confirm the system’s viability and stability. Among them, the accuracy of pallet docking is evaluated 1000 times, and the inaccuracy is kept to a maximum of 6 mm. The recognition rate of pallet recognition is above 99.5% in 7 days of continuous trials.