Research On Technology Of Environment Modelling And Human-robot Interaction For Space Robot Teleoperation

Space exploration is the hotspot of space science research recent years.However,the space environment has the characteristics of microgravity,strong radiation and uncertainty,posing a great security threat to astronauts in the field exploration.Meanwhile,it is still difficult to complete autonomous intelligent robot in a short term.Human-in-Loop based space robot teleoperation is still the main technical means for space exploration at this stage,that is,relying on the ground operators to provide instructions,controlling the space robot to carry out the task,the multi-perception fedback during operation to operators to achieve "immersive" and successfully manipulation.Towards the space teleoperation background,the Virtual Environment Modeling and Human-Robot Interaction technology in the unstructured environment are mainly studies,including the basic framework of space robot teleoperation system based on virtual environment,translational objects dynamic modelling and correction for teleoperation,and haptic constraints dynamic generation method in the point cloud augmented virtual reality environment for teleoperation.This dissertation aims at realizing a space robot teleoperation system with good universality,high system transparency and safe operation space.For the space robot teleoperation ground research,a virtual reality based robot teleoperation system is designed and constructed.Firstly,the kinematics modelling of the haptic device and the SChunk robot is accomplished.A position incremental cumulative method based position mapping method is proposed for a high-precision-large-space SChunk robot manipulation using a small-workspace haptic device.Secondly,a basic virtual environment is implemented including the virtual robot modelling,specific environment modelling,virtual haptic force rendering.Meanwhile,a piecewise linear force space mapping method is proposed,which solves force mapping from a large range virtual force feedback to a small range force output based on the haptic device.Finally,the sensing system and robot control system are realized based on the ROS interface at the slave side.This part provides the theoretical basis and experimental platform for the research and verification of the space teleoperation robot system.In order to realize the robot teleoperation towards the unstructured environment,a virtual environment modelling method based on RGB-D structured point cloud is proposed,including the geometric modelling and the virtual force modelling in the point cloud described environment.First of all,a three-step point cloud registration method for geometric modelling method is proposed.A robot assisted pre-registration is implemented based on the multi coordinates calibration.And a coarse registration method based a novel SURF-Normal feature is realized to improve the matching precision of two point clouds while the Hueassisted ICP algorithm is adopted to further complete the point cloud refinement registration.Meanwhile,for the registrated point cloud,an adaptive sampling algorithm is proposed,the uniformity of point cloud spacing density is realized.Secondly,a collision detection method in point cloud environment based on the bounding ball is proposed for virtual force modelling.Besides a surface normal vector based virtual proxy motion estimation is applied.Finally,the results of the experiment show the effectiveness of the algorithms.It is a challenging task for operators to interact with the remote environment without its dynamic knowledge during teleoperation.A novel system architecture for implementing the translational object modelling and correction during remote interaction is proposed to reconstruct the haptic interaction and predict the object at the local virtual reality based teleoperation.Firstly,the contact dynamic models and friction models are researched,and the system adopts the Mass-Damper-Spring model and the Adapted Karnopp friction model for dynamic modelling in each motion phase.A stress mutation analysis method is proposed for segmenting the translational object motion into static phase,critical phase,and sliding phase.And the static limiting friction is originally estimated in the teleoperation area.Secondly,a novel adaptive forgetting factor Recursive Least Square method and Sliding-block Least Square method based on forgetting factors are studied for high-accuracy parameter estimation.With the estimated model parameters,the motion of the translational object is predicted at the master side.Thirdly,for model consistence between the real and virtual environments,a new correction strategy is used to adaptively update environment model based on JND theory.According to the experimental results,the translational object can be accurately modelled real time,and its motion at the master side can be predicted precisely and corrected promptly.Remote manipulation of a robot without assistance in an unstructured environment is a challenging task for operators.In this paper,a novel methodology for haptic constraints in a point cloud described unstructured environment is proposed to address the human operation limitation.Firstly,the proposed method generates haptic constraints in real time for an unstructured environment,including regional constraints and guidance constraints.A modified implicit surface method is applied for regional constraint generation for the entire point cloud.Additionally,the isosurface derived from the implicit surface is proposed for real-time threedimensional artificial force field estimation.For guidance constraint generation,a new incremental prediction and local artificial force field generation method based on the modified sigmoid model is proposed in an unstructured point cloud described environment.With the generated haptic constraints,the operator can control the robot to realize obstacle avoidance and easily reach the target tasks.Secondly,to overcome the multimode force feedback and the virtual proxy introduced problem in the system,a human-robot sharing control strategy is proposed,which realizes the force constraint changing smoothly based on a fuzzy logic.Finally,the system evaluation is conducted,and the result demonstrates the effectiveness of the proposed method.