Obstacle Avoidance Path Planning for Worm-like Robot Using Bézier Curve

Abstract

Worm-like robots have demonstrated great potential in navigating through environments requiring body shape deformation. Some examples include navigating within a network of pipes, crawling through rubble for search and rescue operations, and medical applications such as endoscopy and colonoscopy. In this work, we developed path planning optimization techniques and obstacle avoidance algorithms for the peristaltic method of locomotion of worm-like robots. Based on our previous path generation study using a modified rapidly exploring random tree (RRT), we have further introduced the Bézier curve to allow more path optimization flexibility. Using Bézier curves, the path planner can explore more areas and gain more flexibility to make the path smoother. We have calculated the obstacle avoidance limitations during turning tests for a six-segment robot with the developed path planning algorithm. Based on the results of our robot simulation, we determined a safe turning clearance distance with a six-body diameter between the robot and the obstacles. When the clearance is less than this value, additional methods such as backward locomotion may need to be applied for paths with high obstacle offset. Furthermore, for a worm-like robot, the paths of subsequent segments will be slightly different than the path of the head segment. Here, we show that as the number of segments increases, the differences between the head path and tail path increase, necessitating greater lateral clearance margins.

Application of the Improved Rapidly Exploring Random Tree Algorithm to an Insect-like Mobile Robot in a Narrow Environment

“…Figure 20 a depicts the trajectory of the RRT algorithm [ 30 ] in location planning for two elevator ports. Although overall path planning is achieved, the randomness of the RRT algorithm prolongs the time required for path exploration [ 31 ].…”

“…The improved algorithm yields smoother and higher-quality paths, making it more suitable for the robot to follow. Figure 20a depicts the trajectory of the RRT algorithm [30] in location planning for two elevator ports. Although overall path planning is achieved, the randomness of the RRT algorithm prolongs the time required for path exploration [31].…”

A worm-inspired robot based on origami structures driven by the magnetic field

“…Equation (6) shows that the pushing force F is proportional to the cross section area A, the square The force analysis on the obstacle is shown in figure 7(c), where G is the gravity, F N is the supporting force, F is the pushing force, and F f is the frictional force. Assume that the mass of the obstacle is about m = 4 g, the coefficient of friction is µ = 0.4, and the angle β = 45 • .…”

“…Worm-like creatures, as the most common soft annelids in moist environment, have attracted tremendous interest for scientists and engineers owing their adaptive locomotion abilities [1,2]. As is well known, worms can navigate effectively in narrow spaces with multi-locomotion patterns such as creeping and crawling [3], rolling [4], turning over [5], and obstacle crossing [6]. The characteristics of worms can certainly shed new light on the design and fabrication of soft robots [7].…”

Moving Object Detection and Tracking by Event Frame from Neuromorphic Vision Sensors

“…Intelligent agents such as robots, unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and autonomous underwater vehicles (AUVs), are widely used in military and civilian fields [1][2][3][4][5][6][7][8]. Object detection and tracking are very important to improve the autonomy of intelligent agents.…”