General introduction to recent advancements in path and trajectory planning for robots and autonomous machines

The main task of mobile robot path planning is: according to the environment model, the mobile robot is based on one or some optimization criteria: such as the minimum work cost, the shortest walking route, the shortest walking time, etc., to find a path in the motion space that does not occur with obstacles. Under the premise of collision, the collision-free path from the starting coordinate point to the target coordinate point allows the robot to reach the destination safely. At present, the path planning methods of mobile robots can be roughly divided into three categories according to their working methods. The first is path planning based on environmental models. It can handle path planning under the condition of fully known obstacle positions and shapes. In the environment, the path planning method based on the environment model will not be applicable. Specific methods such as A * [1], topological graph method [2], etc .; second is local path planning method based on sensor information, typical methods are: artificial potential field method [3], fuzzy logic method [4], etc .; third Behavior-based path planning method [5], which decomposes the navigation problem into independent modules such as collision avoidance and target guidance [6]. Practice shows that it is an effective method to apply neural network to automatic generation of robot trajectory and path planning of mobile robot.

This paper proposes a two-layer path-planning method, where an optimized artificial potential field (APF) method and an improved dynamic window approach (DWA) are used at the global and local layer, respectively. This method enables the robot to plan a better path under a multi-obstacle environment while avoiding the moving obstacles effectively. For the part of global path planning, a new repulsive field is proposed based on the APF method. The length and smoothness of the global path are taken as fitness functions of particle swarm optimization (PSO) to obtain the obstacle influence range, the coefficients of gravitation and repulsion in APF. At the level of local path planning, on the basis of DWA, a fuzzy control scheme is adopted to evaluate the danger level of moving obstacles via collision risk index and relative distance. In brief, compared with existing methods, the robot can reasonably plan a shorter and smoother path with the aid of PSO-based APF, meanwhile quickly react to the moving obstacles and avoid them by fuzzy-based DWA. Finally, a static multi-obstacle environment and two dynamic scenarios with moving obstacles are simulated to verify the effectiveness of the proposed path-planning method.

Two Approaches for Path Planning of Unmanned Aerial Vehicles with Avoidance Zones

A Hybrid Path planning approach combining Artificial Potential Field and Particle Swarm Optimization for Mobile Robot

Aiming at the problems of “local minimum” and “unreachable target” existing in the traditional artificial potential field method in path planning, an improved artificial potential field method was proposed after analyzing the fundamental causes of the above problems. The method solved the problem of local minimum by modifying the direction and influence range of the gravitational field, increasing the virtual target and evaluation function, and the problem of unreachable targets is solved by increasing gravity. In view of the change of motion state of robot fish in amphibious environments, the improved artificial potential field method was fused with a dynamic window algorithm, and a dynamic window evaluation function of the optimal path was designed on the basis of establishing the dynamic equations of land and underwater. Then, the simulation experiment was designed under the environment of Matlab2019a. Firstly, the improved and traditional artificial potential field methods were compared. The results showed that the improved artificial potential field method could solve the above two problems well, shorten the operation time and path length, and have high efficiency. Secondly, the influence of different motion modes on path planning is verified, and the result also reflects that the amphibious robot can avoid obstacles flexibly and reach the target point accurately according to its own motion ability. This paper provides a new way of path planning for the amphibious robot.

Safe, efficient, and feasible path planning for unmanned aerial vehicle (UAV) in complex three-dimensional environments remains a critical challenge in intelligent navigation. Although the traditional artificial potential field (APF) method offers high computational efficiency, it is prone to local minima and target inaccessibility in complex environments. To overcome these limitations, this paper proposes a novel 3D path planning approach that integrates an APF with particle swarm optimization (PSO). The method introduces a Gaussian repulsive function to improve obstacle boundary perception and ensure smooth potential field transitions. Combined with a PSO-based optimization of trajectory points, it effectively enhances the algorithm’s ability to escape local minima and improves global path accessibility. To validate the effectiveness of the proposed method, a comprehensive evaluation framework was developed, incorporating key performance metrics such as total path length, average safety distance, and minimum safety distance. Benchmark comparisons were conducted against several advanced path planning algorithms in static, dynamic, and more complex obstacle environments. In addition, a path executability verification framework was introduced, employing multiple flight controllers to perform trajectory tracking simulations and systematically assess control feasibility and adaptability in complex dynamic scenarios. Experimental results show that the proposed method exhibits clear advantages in both path planning quality and practical executability, particularly demonstrating strong adaptability and robustness in complex environments.

Two-Stage Trajectory Planning for Autonomous Deployment and Target Capture of Space Manipulators

Robots complete tasks in dangerous and extreme environments can significantly enhance people's life safety. Given the complexity and variability of indoor fire accidents, the traditional artificial potential field (APF) method faces challenges in executing the task of path planning. As two types of deep reinforcement learning (DRL) with APF methods, in this paper we leverage an on‐value method deep Q‐learning network (DQN–APF) and an on‐policy method deep deterministic policy gradient (DDPG‐APF) to address the path planning problem. For the purpose of validation upon both the algorithm's efficiency and generalisation, we take Harbin Geographic Information Industrial Park in Heilongjiang Province of China as our research area and incorporate Microsoft HoloLens devices as APF builders for simulating potential fields. Experiments show that both optimisation algorithms can significantly enhance the capabilities on path planning in complex environments. In six random obstacles environments, the DDPG‐APF is better than the DQN–APF method, with a 14.4% higher efficiency. Furthermore, for the generalisation of the algorithm, DDPG‐APF requires less time to plan a more efficient path than DQN–APF, respectively. The experimental results indicate that both optimisation algorithms effectively enable path planning for robots in indoor fire accidents and demonstrate good generalisation abilities.

With the rapid development of new-generation artificial intelligence and Internet of Things technology, mobile robot technology has been widely used in various fields. Among them, the autonomous path-planning technology of mobile robots is one of the cores for realizing their autonomous driving and obstacle avoidance. This study conducts an in-depth discussion on the real-time and dynamic obstacle avoidance capabilities of mobile robot path planning. First, we proposed a preprocessing method for obstacles in the grid map, focusing on the closed processing of the internal space of concave obstacles to ensure the feasibility of the path while effectively reducing the number of grid nodes searched by the A* algorithm, thereby improving path search efficiency. Secondly, in order to achieve static global path planning, this study adopts the A algorithm. However, in practice, algorithm A has problems such as a large number of node traversals, low search efficiency, redundant path nodes, and uneven turning angles. To solve these problems, we optimized the A* algorithm, focusing on optimizing the heuristic function and weight coefficient to reduce the number of node traversals and improve search efficiency. In addition, we use the Bezier curve method to smooth the path and remove redundant nodes, thereby reducing the turning angle. Then, in order to achieve dynamic local path planning, this study adopts the artificial potential field method. However, the artificial potential field method has the problems of unreachable target points and local minima. In order to solve these problems, we optimized the repulsion field so that the target point is at the lowest point of the global energy of the gravitational field and the repulsive field and eliminated the local optimal point. Finally, for the path-planning problem of mobile robots in dynamic environments, this study proposes a hybrid path-planning method based on a combination of the improved A* algorithm and the artificial potential field method. In this study, we not only focus on the efficiency of mobile robot path planning and real-time dynamic obstacle avoidance capabilities but also pay special attention to the symmetry of the final path. By introducing symmetry, we can more intuitively judge whether the path is close to the optimal state. Symmetry is an important criterion for us to evaluate the performance of the final path.

In order to improve the efficiency and effectiveness of path planning for mobile robots in dynamic environment, based on the comprehensive analysis of multiple path planning algorithms, an improved algorithm based on fast-spreading random tree algorithm is proposed. The proposed algorithm effectively combines the artificial potential field(APF) method with the biased target RRT algorithm, and judges the local minimum point in the process of solving the APF method to complete the switching with the biased target RRT algorithm. The motion environment simulation model is established in the Python environment, and the path planning simulation case of the mobile robot in the dynamic environment is completed. The simulation results show that the combined algorithm can avoid the phenomenon that the APF method falls into the local minimum point while taking the global and real-time performance into account, reduce the planning time of the biased target RRT algorithm, and improve the quality of path generation.

Path planning for mobile robots using Bacterial Potential Field for avoiding static and dynamic obstacles

Trajectory planning of robots is not only an important research field of robotics, but also an challenging combination of artificial intelligence and robotics. Trajectory planning is a key technology of Unmanned Aerial Vehicle (UAV). It is also a prerequisite for autonomous navigation of UAV. Under the static environment distributing of obstacles, seeking a shortest path from the starting point to the target point for global path planning has important scientific significance. This paper presents a method for collision-free trajectory planning and design of UAV. Using grids to process the trajectory environment of UAV and then find the shortest path from the initial point to the target point based on rasterization environment using A* algorithm. The simulation of trajectory planning is implemented under the Microsoft Visual C++ 6.0 development environment. From the simulation of trajectory planning, we can clearly prove that UAV can seek a shortest path from the initial point to the target point based on A* algorithm.

This work adopts a standard Denavit–Hartenberg method to model a PUMA 560 spot welding robot as the object of study. The forward and inverse kinematics solutions are then analyzed. To address the shortcomings of the ant colony algorithm, factors from the particle swarm optimization and the genetic algorithm are introduced into this algorithm. Subsequently, the resulting hybrid algorithm and the ant colony algorithm are used to conduct trajectory planning in the shortest path. Experimental data and simulation result show that the hybrid algorithm is significantly better in terms of initial solution speed and optimal solution quality than the ant colony algorithm. The feasibility and effectiveness of the hybrid algorithm in the trajectory planning of a robot are thus verified.

The humanoid robot garners paramount interest because of its ability to mimic human-like behavior in real-time environments. In this paper, a hybridized Artificial Potential Field (APF)-Broyden Fletcher Goldfarb Shanno (BFGS) Quasi-Newton technique is being proposed for the trajectory planning of the humanoid robot. The proposed methodology combines the faster local search BFGS Quasi-Newton method with the global search APF method for an efficient and faster-hybridized trajectory planning technique. The humanoid robot herein is being tested in a multi-humanoid working space having uneven surfaces. The data obtained from the sensors concerning the location of obstacles and the target are at first imparted to the APF method, which supplies an intermediate turning angle dependent on the predesigned training model of the APF method. The obtained turning angle is then fed into the BFGS quasi-newton method to produce an optimum turning angle, which inevitably guides the humanoid robots to the target by keeping a minimum safe distance from the obstacles. Multi-humanoid robots are employed in an environment having random static obstacles, with unique targets for them to reach. The proposal of using a multi-humanoid system arises the chance of inter-collision among the humanoids. To get rid of this situation, the dining philosophers controller is being implemented. The simulations and experiments are carried using the proposed technique to ratify the efficacy of the proposed technique. The experimental and simulation results yield a suitable acceptance rate under 5%. In comparison with the previously used navigational technique, it proves to be comfortably skillful and robust for the trajectory planning of the humanoid robot.

Particle Swarm Optimization Applied to Spacecraft Reentry Trajectory