The rapidly exploring random tree and its variants (RRTs) have been widely adopted as the motion planning algorithms for mobile robots. However, the trap space problem, such as mazes and S-shaped corridors, hinders their planning efficiency. In this article, we present a generalized Voronoi diagram (GVD)-based heuristic path planning algorithm to generate a heuristic path, guide the sampling process of RRTs, and further improve the motion planning efficiency of RRTs. Different from other heuristic algorithms that only work in certain environments or depend on specified parameter setting, the proposed algorithm can automatically identify the environment feature and provide a reasonable heuristic path. First, the given environment is initialized with a lightweight feature extraction from the GVD, which guarantees that any state in the free space can be connected to the feature graph without any collision. Second, to remove the redundancy of feature nodes, a feature matrix is proposed to represent connections among feature nodes and a corresponding feature node fusion technique is utilized to delete the redundant nodes. Third, based on the GVD feature matrix, a heuristic path planning algorithm is presented. This heuristic path is then used to guide the sampling process of RRTs and achieve real-time motion planning. The proposed GVD feature matrix can be also utilized to improve the efficiency of the replanning. Through a series of simulation studies and real-world implementations, it is confirmed that the proposed algorithm achieves better performance in heuristic path planning, feature extraction of free space, and real-time motion planning.

Many kinds of peg-in-hole assembly strategies for an industrial robot have been reported in recent years. Most of these strategies are realized by utilizing visual and force sensors to assist robots. However, complex control algorithms that are based on visual and force sensors will reduce the assembly efficiency of a robot. This issue is thoughtless in traditional assembly strategies but is critical to further improve the efficiency of assembly automation. In this work, a new assembly strategy that is based on a displacement sensor and a variable compliant center is proposed to improve robot performance in assembly tasks. First, an elastic displacement device for this assembly strategy is designed, and its performance is analyzed. The displacement signal generated by the displacement sensor is used to detect the contact state of the peg and hole and to guide the robot to adjust the posture. Second, an assembly strategy, including the advantages of passive compliance and active compliance, and a simple assembly control system are designed to improve the assembly efficiency. Last, the effectiveness of the proposed assembly method is experimentally verified using a robot with 6 degrees of freedom and a chamferless peg and hole with a small clearance (0.1 mm). The experimental results show that the assembly strategy can successfully complete the precision peg-in-hole assembly and assist the robot in accurate assembly in industrial applications.

The safety issue of robot becomes a key challenge in modern manufacturing industry, especially in the scenario of frequent human-robot collaboration. This paper presents a flexible ultrasonic proximity sensing skin which can be integrated on the surface of robot arm to ensure the safe human-robot collaboration. The sensing skin is composed of three pieces of aluminum nitride based piezoelectric micromachined ultrasonic transducers (PMUTs) arrays at the resonance frequency of 115 kHz and chip size of 4 mm $\times 4$ mm $\times0.5$ mm. The three pieces of PMUTs arrays are evenly integrated on the flexible printed circuit board (FPCB) with a device size of 8.8 mm $\times60$ mm $\times1.7$ mm. The proximity sensing skin can detect obstacles within 300 mm and 60 degrees in front, and shows a large sensing coverage in comparison with capacitive/infrared devices. Besides, it is capable of sensing transparent materials like glass. Finally, since the small size and good scalability of micro-electro-mechanical system (MEMS), the PMUTs are easy to install onto robots and to integrate with other ultrasonic transducers in the future. This PMUTs based sensing skin will greatly enhance the sensing capability of robots and human safety during collaborative work.

Autonomous robot exploration has received widespread attention as a crucial step in constructing maps for unknown environments. Among the research results in the field of robotic exploration, the boundary exploration algorithm based on rapidly exploring random tree (RRT) performs well in most scenarios, which guides the robot motion by taking the frontier with the maximum revenue value as the target point discretely. However, this discrete target selection scheme ignores the geometric continuity of the environment, and the robot often turns to another area before completing a geometrically continuous area (e.g., a complete room), which results in the disorderly motion of the robot and a relatively low exploration efficiency. This work proposes a heuristics biased sampling-based robot exploration strategy, which utilizes the semantic information of the environment as the heuristics to guide the robot exploration. First, a lightweight network model is proposed for heuristic object recognition. Second, a geometrically continuous region is estimated based on the heuristic object, and frontiers are then extracted in this region by using a biased sampling method. Finally, the heuristic information gain model is designed to determine the target frontier for the exploration, which instructs the robot to select the frontiers in the heuristic region with a higher priority. In this way, the robot can effectively use the heuristic knowledge of the environment to improve the efficiency of exploration. Compared with the RRT-based exploration methods in simulations and real-world environmental studies, and the experimental results prove the feasibility and effectiveness of our approach.

The sampling-based partial motion planning algorithm has been widely applied in real-time mobile robot navigation for its computational savings and its flexibility in avoiding obstacles. However, in some complex environments, partial planning algorithms are prone to fall into traps, resulting in the failure of motion planning. This paper proposes a feature tree algorithm based on Generalized Voronoi Diagram (GVD) to generate heuristic paths to guide partial motion planning. A GVD feature extraction algorithm is proposed to reduce the redundancy in the representation of obstacle-free regions and improve the searching efficiency in heuristic planning process. The feature node set guarantees that any node from obstacle-free regions can be connected to at least one feature node without any collision. For one map, the feature nodes only need to be extracted once and then can be reused in different scenarios on the same map. Thus, the feature extraction can be executed off-line. Based on GVD feature nodes, a feature tree is reported to generate a heuristic path and the nodes on the heuristic path are utilized sequentially as sub-goals to guide the partial motion planning. When the target changes, the feature tree can quickly replan a new heuristic path. The experimental studies reveal that our proposed method can significantly improve the robot motion planning efficiency and the navigation success rate in trapped environments.