Published on January 3, 2016
slide 1: International Journal of Engineering Research Science IJOER Vol-1 Issue-9 December- 2015 Page | 182 On the Modeling and Simulation of Collision and Collision-Free Motion for Planar Robotic Arm Galia V. Tzvetkova Institute of mechanics Bulgarian Academy of Sciences SOFIA BULGARIA Abstract— Safety functioning is considered as an important issue within overall designing process of autonomous robotic systems. A general structure of collision detection and avoidance system for planar robotic arms is proposed. Simulation results of collision and collision-free motions are presented. Keywords— Collision avoidance planar robotic arms. I. INTRODUCTION Manufacturing service research exploration and many other areas of human activity require a lot of handling operations of a wide variety of objects materials parts etc. The handling operations can be very simple and fuzzy or very complicated and precise. Robotic arms are the right tool to automate these operations and to facilitate the processes in almost of any area. At this stage robotic arms most of them known as manipulation robots work successfully in many manufacturing areas. Usually those areas are static and unchanging work environments which allow the robotic arms to perform their operations according to advanced done programming. The next stage challenge for the robotic arms is their capability to work in considerably cluttered and changing environments that can influence their proper functioning. Besides precise execution of the desired operations the next stage robotic arms must ensure safety functioning within their working space. In cases of dynamic and cluttered environments the robotics arms have to adapt to unknown in advance situations. The most important case to consider is ensuring a safety work—namely avoidance of collisions and accidents with the surrounding objects. The author’s purpose of this paper is to begin a research on problems of safety functioning of autonomous robotic mechanisms in various working environments. II. FUNCTIONAL DESCRIPTION OF COLLISION PREVENTION SYSTEM The collision detection and prevention system consists of three modules that perform the following functions: MODULE 1: “Motion Control” This module implements the motion control according to set performance parameters. The motion control includes moving the arm end-effectors ЕЕ to desired positions of the working space and/or realization of desired trajectories of the arm. MODULE 2:“Detection of Obstacle and Distance Measurement” This module checks for presence of obstacles within the working area of the robotic arm and obtains the necessary measurements such as the global distance between the arm EE and an obstacle as well as additional specific distance measurements thus ensuring the complete measuremnet information for the safety work of the arm. MODULE 3:“Judgment for Situation and Decision for Action This module evaluates the working scene of the arm estimates the degree of danger of the current situation and decides how to safely continue the current process. The possible decisions and actions are: i. No danger situation and the motion can continue to the desired position or along the desired trajectory ii. Collision is possible – correction of the desired position or the desired trajectory iii. STOP - inevitable collision detected impossibility to correct the motion operator assistance required. The block structure of the collision detection and prevention system is presented in Fig.1. slide 2: International Journal of Engineering Research Science IJOER Vol-1 Issue-9 December- 2015 Page | 183 FIG 1. BLOCK STRUCTURE OF THE COLLISION PREVENTION SYSTEM The system variables of the modules are described as follows: q P possitions succesive EE - position point set towards moves 2 link - t 2 q t 2 q position point set towards moves link1 - t 1 q t 1 q 2 controller and 1 controller for points set - L2 L1 EE effector end arm for the point set - EE L2 L1 - 1 1 P ... 1 q P ... 1 1 P Q MODULE g g g g G to 2 link of distance of t measuremen - 3 y to 1 link of distance of t measuremen - 2 y to distance global of t measuremen _ - 1 y of presence for check - 1 F L2_ L1_ OBST 2 2 OBSTACLE 2 OBSTACLES 2 2 obstacle dist obstacle dist dist glob obst obst MODULE slide 3: International Journal of Engineering Research Science IJOER Vol-1 Issue-9 December- 2015 Page | 184 help s operator of necessity motion impossible STOP - - - 3 13 y corection ry trajecto possible is collision - - - 3 12 y motion of on continuati danger collision no - - - 3 11 y danger of degree a for Decision - - - CC - - evaluation - 3 1 Y closeness of criterion - CC 3 1 F closnss OBST closnss 3 glob_dist MODULE III. MATHEMATICAL BASIS The dynamic model of n-dof robotic arm is well known as 1: τ θ θ C θ θ D θ 1 Where n R τ θ θ θ are n-vectors of joints position velocity acceleration and input driving torque n n R D θ matrix of inertia forces n R θ C θ matrix of centrifugal and coriolis forces The state vector n R t 2 x consists of robotic arm joint angles and velocities θ θ : T 2 1 2 1 ... t t t t t t t n n x The equation 1 is transformed in state space form: t t Bu Ax x t 2 where 0 0 I 0 A 1 θ θ C θ D 0 B and the control input is t t τ u . A state space equation for a separate link i can be written as: n i j j j ij i i i i i t l V t u B t x A t x 1 3 with n ... 12 0 0 i x x i i as initial conditions for the links and t l j - the closest distance between link j and detected obstacle denoted as OBST OBST ij V matrix of connections. A. Free of Obstacle Control Equation 2 is presented in discrete form as 2: k k k u B x A x 1 4 The matrixes A and B are calculated according to known formulas: 1 - t - pI 1 - L T 0 0 T A A d HT 0 0 0 T B B 5 and T 0 is discretization interval for the system variables. slide 4: International Journal of Engineering Research Science IJOER Vol-1 Issue-9 December- 2015 Page | 185 Closing the open system 4 by a feedback gain matrix k2 k1 K and after some routine transformations the state of the closed system becomes: G 1 k k k u B x K B A x 6 The control input is calculated as 2: k k k x K G u EE G 7 where G k u is the control input that ensures free movement of the arm end-effectors EE to the desired position when no obstacles exist within the working space of the robotic arm. B. Obstacle consideration and motion correction The closed system is described after taking into account obstacle existence as: obst k k k k Z EE 1 F G B x K B A x 8 The control input contains a second component obst k u : obst k k k u u u G 9 where T 0 0 T 0 F C Φ F d and obst k u is the component of the control vector that realizes trajectory correction after considering existence of obstacle around the EE or the link j . Combining 7 8 and 9 one can get the model: k Z k k k k z F u B Ku B x A x G 1 10 The following relation is considered for the purpose of control correction as a result of existence of potential collision for link j and an obstacle detected in the near closeness: k j k j Z k j l z u F B 11 where k j l is the closest measured distance between the link j and the obstacle. Based on the above equation it is calculated: k j k Z k j l u 1 1 B z F B 12 where Z k j u is the control input to correct the motion because of obstacle presence near link j . IV. BASIC SIMULATIONS FOR 2 DOF PLANAR ROBOTIC ARM The results from a computer program simulation are shown in Fig. 2abcd 3. Fig.2a shows a potential collision situation. The second link is very close to the obstacle. Fig.2b shows collision occurrence when the second link crashes the obstacle. Fig. 2c shows obstacle avoidance from the robotic arm end-effectors. Fig.2d shows collision-free motion after the collision avoidance. slide 5: International Journal of Engineering Research Science IJOER Vol-1 Issue-9 December- 2015 Page | 186 Potential Collision Collision Occurrence Collision Avoidance Fig. 2a Fig.2b Fig.2c Collision-Free Motion Fig. 2d V. CONCLUSION A generic functional model for collision avoidance and collision free-motion of robotic arms is proposed. Basic simulation result for two dof planar robotic arm is presented. Further detailization of research is under planning. REFERENCES 1 M. W. Spong S. Hutchinson and M. Vidyasagar ―Robot Modeling and Control‖ by John Wiley Sons 2006. 2 K. Ogata―System Dynamics‖ 4th Edition Pearson Education Inc. 2004. 3 G.V.Tzvetkova ―Network Modeling of System Goals Attainment‖ Proceedings for the 3rd International Innovations and Real-Time Applications of Distributed Sensor Networks Symposium. Shreveport Convention Center Shreveport Louisiana USA November 26-27 2007.