Abstract: Techniques for predicting positions of objects in an environment are described herein. For example, the techniques may include clustering predicted object positions and using a representative target for one or more clusters to determine a predicted position of an autonomous vehicle. The clusters may vary in size, number of positions, etc. Object positions can be selected from one or more clusters for consideration during vehicle planning, which may include simulation.

Abstract: A machine-learned architecture may predict multiple paths that an object could take in the future without regard to time at which the object may occupy positions identified by one of those paths. These time-invariant paths may be used by an autonomous vehicle to filter detected objects by relevance to an autonomous vehicle's plans, improve prediction of an object's reaction to a vehicle candidate trajectory, determine right-of-way between object(s) and the autonomous vehicle, match detected objects to lanes, and/or improve prediction of odd or out-of-turn object behavior of an object.

Abstract: A machine-learned architecture may predict a set of spatially-diverse paths that an object may take in the future. The paths generated by this architecture may be time-invariant (e.g., not identifying a time at which the object may occupy a position along one of these paths) but can be used by a second machine-learned model to predict progress in time along these paths. This segregation of the spatial paths and progress in time along the paths improves the accuracy of the ultimate prediction and better captures rare object behavior.

Abstract: A heating device for producing carbon fibers from a thread-shaped fiber starting material, wherein the heating device has a central tubular induction heating element through which the fiber starting material is moved. The tubular induction heating element is surrounded by thermal insulation. At least one mid- to high-frequency induction coil is provided outside the thermal insulation, and an inert gas flows through the central induction heating element, in particular, for carbonizing and/or graphitizing the fiber starting material. For energy optimization, a first and a second tube element is provided on the outer side of the thermal insulation. The elements are made of material that is transparent to the induction field of the mid- to high-frequency induction coil and are spaced apart from one another by an annular gap through which the inert gas flows.

Abstract: A heating device for producing carbon fibers from a thread-shaped fiber starting material, wherein the heating device has a central tubular induction heating element through which the fiber starting material is moved. The tubular induction heating element is surrounded by thermal insulation. At least one mid- to high-frequency induction coil is provided outside the thermal insulation, and an inert gas flows through the central induction heating element, in particular, for carbonizing and/or graphitizing the fiber starting material. For energy optimization, a first and a second tube element is provided on the outer side of the thermal insulation. The elements are made of material that is transparent to the induction field of the mid- to high-frequency induction coil and are spaced apart from one another by an annular gap through which the inert gas flows.

Abstract: A precision laser based method of marking semiconductor wafers, packages, substrates or similar workpieces is provided. The workpieces have articles which may include die, chip scale packages, circuit patterns and the like. The marking occurs in a workpiece marking system and within a designated region relative to an article position. The method includes determining at least one location from which reference data is to be obtained using (a) information from which a location of an article is defined, and (b) a vision model of at least a portion of at least one article. Reference data is obtained to locate a feature on a first side of (a second) workpiece using at least one signal from a first sensor. The method further includes positioning a marking field relative to the workpiece so as to position a laser beam at a marking location on a second side of the workpiece. The positioning is based on the feature location.

Abstract: A precision laser based method of marking semiconductor wafers, packages, substrates or similar workpieces is provided. The workpieces have articles which may include die, chip scale packages, circuit patterns and the like. The marking occurs in a workpiece marking system and within a designated region relative to an article position. The method includes determining at least one location from which reference data is to be obtained using (a) information from which a location of an article is defined, and (b) a vision model of at least a portion of at least one article. Reference data is obtained to locate a feature on a first side of (a second) workpiece using at least one signal from a first sensor. The method further includes positioning a marking field relative to the workpiece so as to position a laser beam at a marking location on a second side of the workpiece. The positioning is based on the feature location.

Abstract: A system and method for inspecting machine readable marks on one side of a wafer without requiring transmission of radiant energy from another side of the wafer and through the wafer. The wafer has articles which may include die, chip scale packages, circuit patterns and the like. The marking occurs in a wafer marking system and within a designated region relative to an article position. The articles have a pattern on a first side. The method includes the steps of imaging a first side of the wafer, imaging a second side of the wafer, establishing correspondence between a portion of first side image and a portion of a second side image, and superimposing image data from the first and second sides to determine at least the position of a mark relative to an article.

Abstract: A precision laser based method of marking semiconductor wafers, packages, substrates or similar workpieces is provided. The workpieces have articles which may include die, chip scale packages, circuit patterns and the like. The marking occurs in a workpiece marking system and within a designated region relative to an article position. The method includes determining at least one location from which reference data is to be obtained using (a) information from which a location of an article is defined, and (b) a vision model of at least a portion of at least one article. Reference data is obtained to locate a feature on a first side of (a second) workpiece using at least one signal from a first sensor. The method further includes positioning a marking field relative to the workpiece so as to position a laser beam at a marking location on a second side of the workpiece. The positioning is based on the feature location.

Abstract: A system for semiconductor wafer marking is provided. The system includes: (a) a first positioning subsystem for positioning a laser marking field relative to a wafer, the positioning along a first direction; (b) an alignment vision subsystem; (c) a laser marker including a laser for marking a location within the marking field with a laser marking beam; (d) a calibration program for calibrating at least one subsystem of the system; and (e) a controller. The marking field is substantially smaller than the wafer, and the laser marker includes a scan lens for optically maintaining a spot formed by the beam on the wafer within an acceptable range about the location within the marking field so as to avoid undesirable mark variations associated with wafer sag or other variations in depth within the field.

Abstract: Surfaces of silicon wafers used in making computer chips are marked by dimples engraved by pulses of radiation from a Nd:YAG or Nd:YLF laser pumped by diode lasers (commonly called a DPY laser). A beam of radiation from the DPY laser is focused on and moved across the surface of a wafer so as to produce dimples in a cluster positioned on a microgrid subdividing elements of a dot-matrix representation of a text written in alpha-numeric or bar-code characters.

Abstract: A system and method for inspecting machine readable marks on one side of a wafer without requiring transmission of radiant energy from another side of the wafer and through the wafer. The wafer has articles which may include die, chip scale packages, circuit patterns and the like. The marking occurs in a wafer marking system and within a designated region relative to an article position. The articles have a pattern on a first side. The method includes the steps of imaging a first side of the wafer, imaging a second side of the wafer, establishing correspondence between a portion of first side image and a portion of a second side image, and superimposing image data from the first and second sides to determine at least the position of a mark relative to an article.