Abstract
This paper presents a run-time solver for the inverse kinematics of a robotic arm implemented on a heterogeneous Multi-Processor System-on-Chip (MPSoC). The solver has been formulated as an optimization problem, in which two levels of algorithmic parallelism are proposed: i) the Nelder-Mead derivative-free method used as the optimization engine is modified to allow the evaluation of the cost function in multiple vertices simultaneously, ii) the trajectory is divided into non-overlapping segments, in which all the points are solved concurrently. Algorithmic parallelism is supported by a variable number of parallel instances of a custom hardware accelerator, which speeds up the computation of the forward kinematics equations of the robot required during the resolution of the inverse kinematics. This adaptable scheme provides run-time scalability in terms of trajectory accuracy, logic resources, dependability, and execution time. New design methodologies are used to unify the modeling of the software and hardware partitions of the controller while transparently providing adaptability. They are based on the dataflow Model of Computation (MoC), supported by the PREESM prototyping tool. This tool has been extended to support the use of dynamically reconfigurable hardware accelerators implemented using the ARTICo3 framework. The proposal has been validated with a python-based robotic arm simulator. Experimental results show how the proposed parallelism, combined with hardware acceleration, enables the run-time resolution of the trajectory with adaptable performance using a Xilinx Zynq UltraScale+ MPSoC device.
Highlights
Inverse Kinematics (IK) is a well-known problem in robotics, which aims at determining the joint angles that make the end-effector of the robot reach a given position
Heterogeneous Multi-Processor System-on-Chip (MPSoC) combine in a single device different types of computing fabrics, among which there are Central Processing Units (CPUs), Graphic Processing Unit (GPU) and Field Programmable Gate Array (FPGA) [8]
WORK In this paper, we present a novel scheme for solving the IK problem at run time, targeting heterogeneous MPSoC devices
Summary
Inverse Kinematics (IK) is a well-known problem in robotics, which aims at determining the joint angles that make the end-effector of the robot reach a given position. A novel strategy is proposed to define the initial conditions for each IK problem, which allows executing multiple points of the trajectory in parallel This number can be changed at run time, trading it off with the roughness of the movements and power consumption. Solving the inverse kinematics problem at run time with a high accuracy requires intensive computations, which have to be performed by an embedded system attached to the robot In this regard, we propose using a heterogeneous Multi-Processor System-on-Chip (MPSoC) as the IK controller. Heterogeneous MPSoCs combine in a single device different types of computing fabrics, among which there are Central Processing Units (CPUs), Graphic Processing Unit (GPU) and Field Programmable Gate Array (FPGA) [8] This combination generally results in higher computing performance, lower power consumption, and higher flexibility when compared to homogeneous multi-core solutions.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
