Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
Member:sungbeanJo_paper [2021/03/04 22:29] sungbean |
Member:sungbeanJo_paper [2021/04/21 22:08] (current) sungbean |
||
---|---|---|---|
Line 1: | Line 1: | ||
- | Assume that each observation o = hi;mi comprises an | + | get_config_param active timestamp_mode |
- | image i and a low-dimensional vector m that we refer to as | + | TIME_FROM_INTERNAL_OSC |
- | measurements, following Dosovitskiy and Koltun [9]. The | + | get_config_param active multipurpose_io_mode |
- | controller F is represented by a deep network. The network | + | OUTPUT_OFF |
- | takes the image i, the measurements m, and the command | + | get_config_param active sync_pulse_in_polarity |
- | c as inputs, and produces an action a as its output. The | + | ACTIVE_LOW |
- | action space can be discrete, continuous, or a hybrid of these. | + | get_config_param active nmea_in_polarity |
- | In our driving experiments, the action space is continuous | + | ACTIVE_HIGH |
- | and two-dimensional: steering angle and acceleration. The | + | get_config_param active nmea_baud_rate |
- | acceleration can be negative, which corresponds to braking or | + | BAUD_9600 |
- | driving backwards. The command c is a categorical variable | + | |
- | represented by a one-hot vector. | + | |
- | We study two approaches to incorporating the command | + | |
- | c into the network. The first architecture is illustrated in | + | |
- | Figure 3(a). The network takes the command as an input, | + | |
- | alongside the image and the measurements. These three | + | |
- | inputs are processed independently by three modules: an | + | |
- | image module I(i), a measurement module M(m), and a | + | |
- | command module C(c). The image module is implemented | + | |
- | as a convolutional network, the other two modules as fullyconnected | + | |
- | networks. The outputs of these modules are concatenated | + | |
- | into a joint representation:. | + | |
- | The control module, implemented as a fully-connected network, | ||
- | takes this joint representation and outputs an action | ||
- | A(j). We refer to this architecture as command input. | ||
- | It is applicable to both continuous and discrete commands | ||
- | of arbitrary dimensionality. However, the network is not | ||
- | forced to take the commands into account, which can lead | ||
- | to suboptimal performance in practice. | ||
- | We therefore designed an alternative architecture, shown in | ||
- | Figure 3(b). The image and measurement modules are as described | ||
- | above, but the command module is removed. Instead, | ||
- | we assume a discrete set of commands C = fc0; : : : ; cKg | ||
- | (including a default command c0 corresponding to no specific | ||
- | command given) and introduce a specialist branch Ai | ||
- | for each of the commands ci. The command c acts as a | ||
- | switch that selects which branch is used at any given time. | ||
- | The output of the network is thus | ||
- | We refer to this architecture as branched. The branches Ai | ||
- | are forced to learn sub-policies that correspond to different | ||
- | commands. In a driving scenario, one module might specialize | ||
- | in lane following, another in right turns, and a third in | ||
- | left turns. All modules share the perception stream. |