Sensor Acquisition Geometry. We represent in a the acquisition of a rotating sensor, which is split into 1 ⁄3 turn slices in b. As the Laser emitters position forms an angle of over 17.3 • around the sensor head, taking slices with respect to the sensor rotation θ results in a jagged profile.

Sensor Acquisition Geometry. We represent in a the acquisition of a rotating sensor, which is split into 1 ⁄3 turn slices in b. As the Laser emitters position forms an angle of over 17.3 • around the sensor head, taking slices with respect to the sensor rotation θ results in a jagged profile.

Source publication
Preprint
Full-text available
Roof-mounted spinning LiDAR sensors are widely used by autonomous vehicles, driving the need for real-time processing of 3D point sequences. However, most LiDAR semantic segmentation datasets and algorithms split these acquisitions into $360^\circ$ frames, leading to acquisition latency that is incompatible with realistic real-time applications and...

Context in source publication

Context 1
... the fibers (i.e. the individual lasers) do not all face the same direction: they are arranged around the sensor's heads at different angles, with a range of more than 17.3 • . This means that the points within a packet are not vertically aligned but present a jagged profile as seen in Figure 4. In order to obtain frames with straight edges such as those of SemanticKITTI [3], we would have to consider an acquisition over a sensor rotation of 377 • , adding a further 5ms of latency. ...

Similar publications

Article
Full-text available
Automated driving technology relies heavily on continuous, high-precision, and high-reliability positioning in complex urban situations. With the mass production of autonomous vehicles, the issue of keeping equipment as cheap as possible while maintaining precision is getting much attention. The low-cost global navigation satellite system (GNSS)/mi...
Article
Full-text available
Lane detection is one of the most basic and essential tasks for autonomous vehicles. Therefore, the fast and accurate recognition of lanes has become a hot topic in industry and academia. Deep learning based on a neural network is also a common method for lane detection. However, due to the huge computational burden of the neural network, its real-...
Article
Full-text available
Estimating the distance to objects is crucial for autonomous vehicles, but cost, weight or power constraints sometimes prevent the use of dedicated depth sensors. In this case, the distance has to be estimated from on-board mounted RGB cameras, which is a complex task especially for environments such as natural outdoor landscapes. In this paper, we...
Preprint
Full-text available
Trajectory prediction and behavioral decision-making are two important tasks for autonomous vehicles that require good understanding of the environmental context; behavioral decisions are better made by referring to the outputs of trajectory predictions. However, most current solutions perform these two tasks separately. Therefore, a joint neural n...