The present disclosure discloses a magnetic flexible tactile sensor structure based on a folding magnetization method, which comprises a flexible body containing a permanent magnetic material; the flexible body has a negative Poisson's ratio structure, and its set area has undergone folding magnetization treatment. The present disclosure also discloses a sensor composed of the above-mentioned sensing structure. The sensor provided by the present disclosure can be applied in sealed and wireless scenarios. The present disclosure can detect the size and position of force. The flexible sensor has broad prospects in the application of the touch skin of robots. The function between the magnetic field-based sensing magnet and the Hall element is contactless. In some cases where it is difficult to establish isolation of the connection lines, it can also be used as an unfettered tactile sensor.

Systems and methods for detecting pose and force against an object are provided. A method includes receiving a signal from a deformable sensor comprising data from a deformation region in a deformable membrane resulting from contact with the object utilizing an internal sensor disposed within an enclosure and having a field of view directed through a medium and toward a bottom surface of the deformable membrane. The method also determines a pose of the object based on the deformation region of the deformable membrane. The method also determines an amount of force applied between the deformable membrane and the object is determined based on the deformation region of the deformable membrane.

Systems and methods for an electronic skin based robotic system including a robotic interface and a human subject are provided. An e-skin may be applied to the robotic interface. The e-skin applied to the robotic interface may include a plurality of physicochemical sensors. An e-skin may also be applied to the human subject. The e-skin may include electrodes for sensing muscular contractions associated with hand and arm movements as well as electrodes for stimulation. Machine learning techniques may enable decoding of signals to control the robotic hand and arm. The robotic hand and arm may be controlled to approach unknown compounds that may be hazardous. The sensors making up the physicochemical sensors on the e-skin on the robotic hand and arm may include tactile, pressure, temperature, and chemical sensors, as well as other useful sensors. These sensors may enable detection of explosives, organophosphates, pathogenic proteins, and other hazardous compounds.

A control system, method and computer program product cooperate to assist control for an autonomous robot. An interface receives recognition information from an autonomous robot, said recognition information including candidate target objects to interact with the autonomous robot. A display control unit causes a display image to be displayed on a display of candidate target objects, wherein at least two of the candidate target objects are displayed with an associated indication of a target object score.

Systems and methods for detecting pose and force against an object are provided. A method includes receiving a signal from a deformable sensor comprising data from a deformation region in a deformable membrane resulting from contact with the object utilizing an internal sensor disposed within an enclosure and having a field of view directed through a medium and toward a bottom surface of the deformable membrane. The method also determines a pose of the object based on the deformation region of the deformable membrane. The method also determines an amount of force applied between the deformable membrane and the object is determined based on the deformation region of the deformable membrane.

Various tactile sensors and associated methods are enabled. For instance, a sensing apparatus comprises a photosensitive sensor. A compound-eye structure is on the photosensitive sensor and an elastomer layer is on the compound-eye structure. A reflective layer is on the elastomer layer, opposite the compound-eye structure and a light source emits light between the reflective layer and the compound-eye structure.

A method for controlling a robot to insert an object into an insertion. The method includes controlling the robot to hold the object, generating an estimate of a target position to insert the object into the insertion, controlling the robot to move to the estimated target position, taking a camera image using a camera mounted on the robot after having controlled the robot to move to the estimated target position, feeding the camera image into a neural network which is trained to derive, from camera images, movement vectors which specify movements from the positions at which the camera images are taken to insert objects into insertions and controlling the robot to move according to the movement vector derived by the neural network from the camera image.

Optical tactile sensors are provided that include a scaffolding structure, a transparent elastomer material covering at least an end portion of the scaffolding structure, and one or multiple cameras situated on the end portion of the scaffolding structure and embedded within the transparent elastomer material, wherein the one or multiple cameras are situated so as to provide an extended, e.g., up to 360°, field of view about the end portion of the scaffolding structure.

A tactile sensor, including: a contact part (1), configured to receive pressure transmitted when a manipulator grabs a target object, and transmit the pressure to a sensing part (2); the sensing part (2), located on one side of the contact part, the sensing part (2) moving in a direction away from the contact part (1) under an action of the pressure; and a detection part (4), located on a side of the sensing part (2) away from the contact part (1), the detection part (4) sensing a position change of the sensing part (2) to generate a sensing signal, wherein the sensing signal is configured to determine a contact parameter between the manipulator and the target object.

In variants, a method for robot control can include: receiving sensor data of a scene, modeling the physical objects within the scene, determining a set of potential grasp configurations for grasping a physical object within the scene, determining a reach behavior based on the potential grasp configuration, determining a trajectory for the reach behavior, and grasping the object using the trajectory.